Re-exploring the structure of Chinese character images

Introduction

In this notebook we show information retrieval and clustering
techniques over images of Unicode collection of Chinese characters. Here
is the outline of notebook’s exposition:

  1. Get Chinese character images.
  2. Cluster “image vectors” and demonstrate that the obtained
    clusters have certain explainability elements.
  3. Apply Latent Semantic Analysis (LSA) workflow to the character
    set.
  4. Show visual thesaurus through a recommender system. (That uses
    Cosine similarity.)
  5. Discuss graph and hierarchical clustering using LSA matrix
    factors.
  6. Demonstrate approximation of “unseen” character images with an
    image basis obtained through LSA over a small set of (simple)
    images.
  7. Redo character approximation with more “interpretable” image
    basis.

Remark: This notebook started as an (extended)
comment for the Community discussion “Exploring
structure of Chinese characters through image processing”
, [SH1].
(Hence the title.)

Get Chinese character images

This code is a copy of the code in the original
Community post by Silvia Hao
, [SH1]:

0zu4hv95x0jjf
Module[{fsize = 50, width = 64, height = 64}, 
  lsCharIDs = Map[FromCharacterCode[#, "Unicode"] &, 16^^4E00 - 1 + Range[width height]]; 
 ]
charPage = Module[{fsize = 50, width = 64, height = 64}, 
    16^^4E00 - 1 + Range[width height] // pipe[
      FromCharacterCode[#, "Unicode"] & 
      , Characters, Partition[#, width] & 
      , Grid[#, Background -> Black, Spacings -> {0, 0}, ItemSize -> {1.5, 1.2}, Alignment -> {Center, Center}, Frame -> All, FrameStyle -> Directive[Red, AbsoluteThickness[3 \[Lambda]]]] & 
      , Style[#, White, fsize, FontFamily -> "Source Han Sans CN", FontWeight -> "ExtraLight"] & 
      , Rasterize[#, Background -> Black] & 
     ] 
   ];
chargrid = charPage // ColorDistance[#, Red] & // Image[#, "Byte"] & // Sign //Erosion[#, 5] &;
lmat = chargrid // MorphologicalComponents[#, Method -> "BoundingBox", CornerNeighbors -> False] &;
chars = ComponentMeasurements[{charPage // ColorConvert[#, "Grayscale"] &, lmat}, "MaskedImage", #Width > 10 &] // Values // Map@RemoveAlphaChannel;
chars = Module[{size = chars // Map@ImageDimensions // Max}, ImageCrop[#, {size, size}] & /@ chars];

Here is a sample of the obtained images:

SeedRandom[33];
RandomSample[chars, 5]
1jy9voh5c01lt

Vector representation of
images

Define a function that represents an image into a linear vector space
(of pixels):

Clear[ImageToVector];
ImageToVector[img_Image] := Flatten[ImageData[ColorConvert[img, "Grayscale"]]];
ImageToVector[img_Image, imgSize_] := Flatten[ImageData[ColorConvert[ImageResize[img, imgSize], "Grayscale"]]];
ImageToVector[___] := $Failed;

Show how vector represented images look like:

Table[BlockRandom[
   img = RandomChoice[chars]; 
   ListPlot[ImageToVector[img], Filling -> Axis, PlotRange -> All, PlotLabel -> img, ImageSize -> Medium, AspectRatio -> 1/6], 
   RandomSeeding -> rs], {rs, {33, 998}}]
0cobk7b0m9xcn
\[AliasDelimiter]

Data preparation

In this section we represent the images into a linear vector space.
(In which each pixel is a basis vector.)

Make an association with images:

aCImages = AssociationThread[lsCharIDs -> chars];
Length[aCImages]

(*4096*)

Make flat vectors with the images:

AbsoluteTiming[
  aCImageVecs = ParallelMap[ImageToVector, aCImages]; 
 ]

(*{0.998162, Null}*)

Do matrix plots a random sample of the image vectors:

SeedRandom[32];
MatrixPlot[Partition[#, ImageDimensions[aCImages[[1]]][[2]]]] & /@ RandomSample[aCImageVecs, 6]
07tn6wh5t97j4

Clustering over the image
vectors

In this section we cluster “image vectors” and demonstrate that the
obtained clusters have certain explainability elements. Expected Chinese
character radicals are observed using image multiplication.

Cluster the image vectors and show summary of the clusters
lengths:

SparseArray[Values@aCImageVecs]
1n5cwcrgj2d3m
SeedRandom[334];
AbsoluteTiming[
  lsClusters = FindClusters[SparseArray[Values@aCImageVecs] -> Keys[aCImageVecs], 35, Method -> {"KMeans"}]; 
 ]
Length@lsClusters
ResourceFunction["RecordsSummary"][Length /@ lsClusters]

(*{24.6383, Null}*)

(*35*)
0lvt8mcfzpvhg

For each cluster:

  • Take 30 different small samples of 7 images
  • Multiply the images in each small sample
  • Show three “most black” the multiplication results
SeedRandom[33];
Table[i -> TakeLargestBy[Table[ImageMultiply @@ RandomSample[KeyTake[aCImages, lsClusters[[i]]], UpTo[7]], 30], Total@ImageToVector[#] &, 3], {i, Length[lsClusters]}]
0erc719h7lnzi

Remark: We can see that the clustering above
produced “semantic” clusters – most of the multiplied images show
meaningful Chinese characters radicals and their “expected
positions.”

Here is one of the clusters with the radical “mouth”:

KeyTake[aCImages, lsClusters[[26]]]
131vpq9dabrjo

LSAMon application

In this section we apply the “standard” LSA workflow, [AA1, AA4].

Make a matrix with named rows and columns from the image vectors:

mat = ToSSparseMatrix[SparseArray[Values@aCImageVecs], "RowNames" -> Keys[aCImageVecs], "ColumnNames" -> Automatic]
0jdmyfb9rsobz

The following Latent Semantic Analysis (LSA) monadic pipeline is used
in [AA2, AA2]:

SeedRandom[77];
AbsoluteTiming[
  lsaAllObj = 
    LSAMonUnit[]\[DoubleLongRightArrow]
     LSAMonSetDocumentTermMatrix[mat]\[DoubleLongRightArrow]
     LSAMonApplyTermWeightFunctions["None", "None", "Cosine"]\[DoubleLongRightArrow]
     LSAMonExtractTopics["NumberOfTopics" -> 60, Method -> "SVD", "MaxSteps" -> 15, "MinNumberOfDocumentsPerTerm" -> 0]\[DoubleLongRightArrow]
     LSAMonNormalizeMatrixProduct[Normalized -> Right]\[DoubleLongRightArrow]
     LSAMonEcho[Style["Obtained basis:", Bold, Purple]]\[DoubleLongRightArrow]
     LSAMonEchoFunctionContext[ImageAdjust[Image[Partition[#, ImageDimensions[aCImages[[1]]][[1]]]]] & /@SparseArray[#H] &]; 
 ]
088nutsaye7yl
0j7joulwrnj30
(*{7.60828, Null}*)

Remark: LSAMon’s corresponding theory and design are
discussed in [AA1, AA4]:

Get the representation matrix:

W2 = lsaAllObj\[DoubleLongRightArrow]LSAMonNormalizeMatrixProduct[Normalized -> Right]\[DoubleLongRightArrow]LSAMonTakeW
1nno5c4wmc83q

Get the topics matrix:

H = lsaAllObj\[DoubleLongRightArrow]LSAMonNormalizeMatrixProduct[Normalized -> Right]\[DoubleLongRightArrow]LSAMonTakeH
1gtqe0ihshi9s

Cluster the reduced dimension
representations
and show summary of the clusters
lengths:

AbsoluteTiming[
  lsClusters = FindClusters[Normal[SparseArray[W2]] -> RowNames[W2], 40, Method -> {"KMeans"}]; 
 ]
Length@lsClusters
ResourceFunction["RecordsSummary"][Length /@ lsClusters]

(*{2.33331, Null}*)

(*40*)
1bu5h88uiet3e

Show cluster interpretations:

AbsoluteTiming[aAutoRadicals = Association@Table[i -> TakeLargestBy[Table[ImageMultiply @@ RandomSample[KeyTake[aCImages, lsClusters[[i]]], UpTo[8]], 30], Total@ImageToVector[#] &, 3], {i, Length[lsClusters]}]; 
 ]
aAutoRadicals

(*{0.878406, Null}*)
05re59k8t4u4u

Using FeatureExtraction

I experimented with clustering and approximation using WL’s function
FeatureExtraction.
Result are fairly similar as the above; timings a different (a few times
slower.)

Visual thesaurus

In this section we use Cosine similarity to find visual nearest
neighbors of Chinese character images.

matPixels = WeightTermsOfSSparseMatrix[lsaAllObj\[DoubleLongRightArrow]LSAMonTakeWeightedDocumentTermMatrix, "IDF", "None", "Cosine"];
matTopics = WeightTermsOfSSparseMatrix[lsaAllObj\[DoubleLongRightArrow]LSAMonNormalizeMatrixProduct[Normalized -> Left]\[DoubleLongRightArrow]LSAMonTakeW, "None", "None", "Cosine"];
smrObj = SMRMonUnit[]\[DoubleLongRightArrow]SMRMonCreate[<|"Topic" -> matTopics, "Pixel" -> matPixels|>];

Consider the character “團”:

aCImages["團"]
0pi2u9ejqv9wd

Here are the nearest neighbors for that character found by using both
image topics and image pixels:

(*focusItem=RandomChoice[Keys@aCImages];*)
  focusItem = {"團", "仼", "呔"}[[1]]; 
   smrObj\[DoubleLongRightArrow]
     SMRMonEcho[Style["Nearest neighbors by pixel topics:", Bold, Purple]]\[DoubleLongRightArrow]
     SMRMonSetTagTypeWeights[<|"Topic" -> 1, "Pixel" -> 0|>]\[DoubleLongRightArrow]
     SMRMonRecommend[focusItem, 8, "RemoveHistory" -> False]\[DoubleLongRightArrow]
     SMRMonEchoValue\[DoubleLongRightArrow]
     SMRMonEchoFunctionValue[AssociationThread[Values@KeyTake[aCImages, Keys[#]], Values[#]] &]\[DoubleLongRightArrow]
     SMRMonEcho[Style["Nearest neighbors by pixels:", Bold, Purple]]\[DoubleLongRightArrow]
     SMRMonSetTagTypeWeights[<|"Topic" -> 0, "Pixel" -> 1|>]\[DoubleLongRightArrow]
     SMRMonRecommend[focusItem, 8, "RemoveHistory" -> False]\[DoubleLongRightArrow]
     SMRMonEchoFunctionValue[AssociationThread[Values@KeyTake[aCImages, Keys[#]], Values[#]] &];
1l9yz2e8pvlyl
03bc668vzyh4v
00ecjkyzm4e2s
1wsyx76kjba1g
18wdi99m1k99j

Remark: Of course, in the recommender pipeline above
we can use both pixels and pixels topics. (With their contributions
being weighted.)

Graph clustering

In this section we demonstrate the use of graph communities to find
similar groups of Chinese characters.

Here we take a sub-matrix of the reduced dimension matrix computed
above:

W = lsaAllObj\[DoubleLongRightArrow]LSAMonNormalizeMatrixProduct[Normalized -> Right]\[DoubleLongRightArrow]LSAMonTakeW;

Here we find the similarity matrix between the characters and remove
entries corresponding to “small” similarities:

matSym = Clip[W . Transpose[W], {0.78, 1}, {0, 1}];

Here we plot the obtained (clipped) similarity matrix:

MatrixPlot[matSym]
1nvdb26265li6

Here we:

  • Take array rules of the sparse similarity matrix
  • Drop the rules corresponding to the diagonal elements
  • Convert the keys of rules into uni-directed graph edges
  • Make the corresponding graph
  • Find graph’s connected components
  • Show the number of connected components
  • Show a tally of the number of nodes in the components
gr = Graph[UndirectedEdge @@@ DeleteCases[Union[Sort /@ Keys[SSparseMatrixAssociation[matSym]]], {x_, x_}]];
lsComps = ConnectedComponents[gr];
Length[lsComps]
ReverseSortBy[Tally[Length /@ lsComps], First]

(*138*)

(*{{1839, 1}, {31, 1}, {27, 1}, {16, 1}, {11, 2}, {9, 2}, {8, 1}, {7, 1}, {6, 5}, {5, 3}, {4, 8}, {3, 14}, {2, 98}}*)

Here we demonstrate the clusters of Chinese characters make
sense:

aPrettyRules = Dispatch[Map[# -> Style[#, FontSize -> 36] &, Keys[aCImages]]]; CommunityGraphPlot[Subgraph[gr, TakeLargestBy[lsComps, Length, 10][[2]]], Method -> "SpringElectrical", VertexLabels -> Placed["Name", Above],AspectRatio -> 1, ImageSize -> 1000] /. aPrettyRules
1c0w4uhnyn2jx

Remark: By careful observation of the clusters and
graph connections we can convince ourselves that the similarities are
based on pictorial sub-elements (i.e. radicals) of the characters.

Hierarchical clustering

In this section we apply hierarchical clustering to the reduced
dimension representation of the Chinese character images.

Here we pick a cluster:

lsFocusIDs = lsClusters[[12]];
Magnify[ImageCollage[Values[KeyTake[aCImages, lsFocusIDs]]], 0.4]
14cnicsw2rvrt

Here is how we can make a dendrogram plot (not that useful here):

(*smat=W2\[LeftDoubleBracket]lsClusters\[LeftDoubleBracket]13\[RightDoubleBracket],All\[RightDoubleBracket];
Dendrogram[Thread[Normal[SparseArray[smat]]->Map[Style[#,FontSize->16]&,RowNames[smat]]],Top,DistanceFunction->EuclideanDistance]*)

Here is a heat-map plot with hierarchical clustering dendrogram (with
tool-tips):

gr = HeatmapPlot[W2[[lsFocusIDs, All]], DistanceFunction -> {CosineDistance, None}, Dendrogram -> {True, False}];
gr /. Map[# -> Tooltip[Style[#, FontSize -> 16], Style[#, Bold, FontSize -> 36]] &, lsFocusIDs]
0vz82un57054q

Remark: The plot above has tooltips with larger
character images.

Representing
all characters with smaller set of basic ones

In this section we demonstrate that a relatively small set of simpler
Chinese character images can be used to represent (or approxumate) the
rest of the images.

Remark: We use the following heuristic: the simpler
Chinese characters have the smallest amount of white pixels.

Obtain a training set of images – that are the darkest – and show a
sample of that set :

{trainingInds, testingInds} = TakeDrop[Keys[SortBy[aCImages, Total[ImageToVector[#]] &]], 800];
SeedRandom[3];
RandomSample[KeyTake[aCImages, trainingInds], 12]
10275rv8gn1qt

Show all training characters with an image collage:

Magnify[ImageCollage[Values[KeyTake[aCImages, trainingInds]], Background -> Gray, ImagePadding -> 1], 0.4]
049bs0w0x26jw

Apply LSA monadic pipeline with the training characters only:

SeedRandom[77];
AbsoluteTiming[
  lsaPartialObj = 
    LSAMonUnit[]\[DoubleLongRightArrow]
     LSAMonSetDocumentTermMatrix[SparseArray[Values@KeyTake[aCImageVecs, trainingInds]]]\[DoubleLongRightArrow]
     LSAMonApplyTermWeightFunctions["None", "None", "Cosine"]\[DoubleLongRightArrow]
     LSAMonExtractTopics["NumberOfTopics" -> 80, Method -> "SVD", "MaxSteps" -> 120, "MinNumberOfDocumentsPerTerm" -> 0]\[DoubleLongRightArrow]
     LSAMonNormalizeMatrixProduct[Normalized -> Right]\[DoubleLongRightArrow]
     LSAMonEcho[Style["Obtained basis:", Bold, Purple]]\[DoubleLongRightArrow]
     LSAMonEchoFunctionContext[ImageAdjust[Image[Partition[#, ImageDimensions[aCImages[[1]]][[1]]]]] & /@SparseArray[#H] &]; 
 ]
0i509m9n2d2p8
1raokwq750nyi
(*{0.826489, Null}*)

Get the matrix and basis interpretation of the extracted image
topics:

H = 
   lsaPartialObj\[DoubleLongRightArrow]
    LSAMonNormalizeMatrixProduct[Normalized -> Right]\[DoubleLongRightArrow]
    LSAMonTakeH;
lsBasis = ImageAdjust[Image[Partition[#, ImageDimensions[aCImages[[1]]][[1]]]]] & /@ SparseArray[H];

Approximation of “unseen”
characters

Pick a Chinese character image as a target image and pre-process
it:

ind = RandomChoice[testingInds];
imgTest = aCImages[ind];
matImageTest = ToSSparseMatrix[SparseArray@List@ImageToVector[imgTest, ImageDimensions[aCImages[[1]]]], "RowNames" -> Automatic, "ColumnNames" -> Automatic];
imgTest
15qkrj0nw08mv

Find its representation with the chosen feature extractor (LSAMon
object here):

matReprsentation = lsaPartialObj\[DoubleLongRightArrow]LSAMonRepresentByTopics[matImageTest]\[DoubleLongRightArrow]LSAMonTakeValue;
lsCoeff = Normal@SparseArray[matReprsentation[[1, All]]];
ListPlot[MapIndexed[Tooltip[#1, lsBasis[[#2[[1]]]]] &, lsCoeff], Filling -> Axis, PlotRange -> All]
0cn7ty6zf3mgo

Show representation coefficients outliers:

lsBasis[[OutlierPosition[Abs[lsCoeff], TopOutliers@*HampelIdentifierParameters]]]
1w6jkhdpxlxw8

Show the interpretation of the found representation:

vecReprsentation = lsCoeff . SparseArray[H];
reprImg = Image[Unitize@Clip[#, {0.45, 1}, {0, 1}] &@Rescale[Partition[vecReprsentation, ImageDimensions[aCImages[[1]]][[1]]]]];
{reprImg, imgTest}
0c84q1hscjubu

See the closest characters using image distances:

KeyMap[# /. aCImages &, TakeSmallest[ImageDistance[reprImg, #] & /@ aCImages, 4]]
1vtcw1dhzlet5

Remark: By applying the approximation procedure to
all characters in testing set we can convince ourselves that small,
training set provides good retrieval. (Not done here.)

Finding more interpretable
bases

In this section we show how to use LSA workflow with Non-Negative
Matrix Factorization (NNMF)
over an image set extended with already
extracted “topic” images.

Cleaner automatic radicals

aAutoRadicals2 = Map[Dilation[Binarize[DeleteSmallComponents[#]], 0.5] &, First /@ aAutoRadicals]
10eg2eaajgiit

Here we take an image union in order to remove the “duplicated”
radicals:

aAutoRadicals3 = AssociationThread[Range[Length[#]], #] &@Union[Values[aAutoRadicals2], SameTest -> (ImageDistance[#1, #2] < 14.5 &)]
1t09xi5nlycaw

LSAMon pipeline with NNMF

Make a matrix with named rows and columns from the image vectors:

mat1 = ToSSparseMatrix[SparseArray[Values@aCImageVecs], "RowNames" -> Keys[aCImageVecs], "ColumnNames" -> Automatic]
0np1umfcks9hm

Enhance the matrix with radicals instances:

mat2 = ToSSparseMatrix[SparseArray[Join @@ Map[Table[ImageToVector[#], 100] &, Values[aAutoRadicals3]]], "RowNames" -> Automatic, "ColumnNames" -> Automatic];
mat3 = RowBind[mat1, mat2];

Apply the LSAMon workflow pipeline with NNMF for topic
extraction:

SeedRandom[77];
AbsoluteTiming[
  lsaAllExtendedObj = 
    LSAMonUnit[]\[DoubleLongRightArrow]
     LSAMonSetDocumentTermMatrix[mat3]\[DoubleLongRightArrow]
     LSAMonApplyTermWeightFunctions["None", "None", "Cosine"]\[DoubleLongRightArrow]
     LSAMonExtractTopics["NumberOfTopics" -> 60, Method -> "NNMF", "MaxSteps" -> 15, "MinNumberOfDocumentsPerTerm" -> 0]\[DoubleLongRightArrow]
     LSAMonNormalizeMatrixProduct[Normalized -> Right]\[DoubleLongRightArrow]
     LSAMonEcho[Style["Obtained basis:", Bold, Purple]]\[DoubleLongRightArrow]
     LSAMonEchoFunctionContext[ImageAdjust[Image[Partition[#, ImageDimensions[aCImages[[1]]][[1]]]]] & /@SparseArray[#H] &]; 
 ]
1mc1fa16ylzcu
1c6p7pzemk6qx
(*{155.289, Null}*)

Remark: Note that NNMF “found” the interpretable
radical images we enhanced the original image set with.

Get the matrix and basis interpretation of the extracted image
topics:

H = 
   lsaAllExtendedObj\[DoubleLongRightArrow]
    LSAMonNormalizeMatrixProduct[Normalized -> Right]\[DoubleLongRightArrow]
    LSAMonTakeH;
lsBasis = ImageAdjust[Image[Partition[#, ImageDimensions[aCImages[[1]]][[1]]]]] & /@ SparseArray[H];

Approximation

Pick a Chinese character image as a target image and pre-process
it:

SeedRandom[43];
ind = RandomChoice[testingInds];
imgTest = aCImages[ind];
matImageTest = ToSSparseMatrix[SparseArray@List@ImageToVector[imgTest, ImageDimensions[aCImages[[1]]]], "RowNames" -> Automatic, "ColumnNames" -> Automatic];
imgTest
1h2aitm71mnl5

Find its representation with the chosen feature extractor (LSAMon
object here):

matReprsentation = lsaAllExtendedObj\[DoubleLongRightArrow]LSAMonRepresentByTopics[matImageTest]\[DoubleLongRightArrow]LSAMonTakeValue;
lsCoeff = Normal@SparseArray[matReprsentation[[1, All]]];
ListPlot[MapIndexed[Tooltip[#1, lsBasis[[#2[[1]]]]] &, lsCoeff], Filling -> Axis, PlotRange -> All]
084vbifk2zvi3

Show representation coefficients outliers:

lsBasis[[OutlierPosition[Abs[lsCoeff], TopOutliers@*QuartileIdentifierParameters]]]
06xq4p3k31fzt

Remark: Note that expected
radical images are in the outliers.

Show the interpretation of the found representation:

vecReprsentation = lsCoeff . SparseArray[H];
reprImg = Image[Unitize@Clip[#, {0.45, 1}, {0, 1}] &@Rescale[Partition[vecReprsentation, ImageDimensions[aCImages[[1]]][[1]]]]];
{reprImg, imgTest}
01xeidbc9qme6

See the closest characters using image distances:

KeyMap[# /. aCImages &, TakeSmallest[ImageDistance[reprImg, #] & /@ aCImages, 4]]
1mrut9izhycrn

Setup

Import["https://raw.githubusercontent.com/antononcube/MathematicaForPrediction/master/MonadicProgramming/MonadicLatentSemanticAnalysis.m"];
Import["https://raw.githubusercontent.com/antononcube/MathematicaForPrediction/master/MonadicProgramming/MonadicSparseMatrixRecommender.m"];
Import["https://raw.githubusercontent.com/antononcube/MathematicaForPrediction/master/Misc/HeatmapPlot.m"]

References

[SH1] Silvia Hao, “Exploring
structure of Chinese characters through image processing”
, (2022),
Wolfram Community.

[AA1] Anton Antonov, “A monad for
Latent Semantic Analysis workflows”
, (2019), Wolfram Community.

[AA2] Anton Antonov, “LSA methods
comparison over random mandalas deconstruction – WL”
, (2022), Wolfram Community.

[AA3] Anton Antonov, “Bethlehem
stars: classifying randomly generated mandalas”
, (2020), Wolfram Community.

[AA4] Anton Antonov, “Random mandalas deconstruction in R, Python, and Mathematica”, (2022), MathematicaForPrediction at WordPress.

[AAp1] Anton Antonov, LSAMon
for Image Collections Mathematica package
, (2022), MathematicaForPrediction
at GitHub
.

Random mandalas deconstruction in R, Python, and Mathematica

Today (2022-02-28) I gave a presentation Greater Boston useR Meetup titled “Random mandalas deconstruction with R, Python, and Mathematica”. (Link to the video recording.)


Here is the abstract:

In this presentation we discuss the application of different dimension reduction algorithms over collections of random mandalas. We discuss and compare the derived image bases and show how those bases explain the underlying collection structure. The presented techniques and insights (1) are applicable to any collection of images, and (2) can be included in larger, more complicated machine learning workflows. The former is demonstrated with a handwritten digits recognition
application; the latter with the generation of random Bethlehem stars. The (parallel) walk-through of the core demonstration is in all three programming languages: Mathematica, Python, and R.


Here is the related RStudio project: “RandomMandalasDeconstruction”.

Here is a link to the R-computations notebook converted to HTML: “LSA methods comparison in R”.

The Mathematica notebooks are placed in project’s folder “notebooks-WL”.


See the work plan status in the org-mode file “Random-mandalas-deconstruction-presentation-work-plan.org”.

Here is the mind-map for the presentation:


The comparison workflow implemented in the notebooks of this project is summarized in the following flow chart:

Random mandalas deconstruction workflow


References

Articles

[AA1] Anton Antonov, “Comparison of dimension reduction algorithms over mandala images generation”, (2017), MathematicaForPrediction at WordPress.

[AA2] Anton Antonov, “Handwritten digits recognition by matrix factorization”, (2016), MathematicaForPrediction at WordPress.

Mathematica packages and repository functions

[AAp1] Anton Antonov, Monadic Latent Semantic Analysis Mathematica package, (2017), MathematicaForPrediction at GitHub/antononcube.

[AAf1] Anton Antonov, NonNegativeMatrixFactorization, (2019), Wolfram Function Repository.

[AAf2] Anton Antonov, IndependentComponentAnalysis, (2019), Wolfram Function Repository.

[AAf3] Anton Antonov, RandomMandala, (2019), Wolfram Function Repository.

Python packages

[AAp2] Anton Antonov, LatentSemanticAnalyzer Python package (2021), PyPI.org.

[AAp3] Anton Antonov, Random Mandala Python package, (2021), PyPI.org.

R packages

[AAp4] Anton Antonov, Latent Semantic Analysis Monad R package, (2019), R-packages at GitHub/antononcube.

Generation of Random Bethlehem Stars

Introduction

This document/notebook is inspired by the Mathematica Stack Exchange (MSE) question “Plotting the Star of Bethlehem”, [MSE1]. That MSE question requests efficient and fast plotting of a certain mathematical function that (maybe) looks like the Star of Bethlehem, [Wk1]. Instead of doing what the author of the questions suggests, I decided to use a generative art program and workflows from three of most important Machine Learning (ML) sub-cultures: Latent Semantic Analysis, Recommendations, and Classification.

Although we discuss making of Bethlehem Star-like images, the ML workflows and corresponding code presented in this document/notebook have general applicability – in many situations we have to make classifiers based on data that has to be “feature engineered” through pipeline of several types of ML transformative workflows and that feature engineering requires multiple iterations of re-examinations and tuning in order to achieve the set goals.

The document/notebook is structured as follows:

  1. Target Bethlehem Star images
  2. Simplistic approach
  3. Elaborated approach outline
  4. Sections that follow through elaborated approach outline:
    1. Data generation
    2. Feature extraction
    3. Recommender creation
    4. Classifier creation and utilization experiments

(This document/notebook is a “raw” chapter for the book “Simplified Machine Learning Workflows”, [AAr3].)

Target images

Here are the images taken from [MSE1] that we consider to be “Bethlehem Stars” in this document/notebook:

imgStar1 = Import["https://i.stack.imgur.com/qmmOw.png"];
imgStar2 = Import["https://i.stack.imgur.com/5gtsS.png"];
Row[{imgStar1, Spacer[5], imgStar2}]
00dxgln7hhmjl

We notice that similar images can be obtained using the Wolfram Function Repository (WFR) function RandomMandala, [AAr1]. Here are a dozen examples:

SeedRandom[5];
Multicolumn[Table[MandalaToWhiterImage@ResourceFunction["RandomMandala"]["RotationalSymmetryOrder" -> 2, "NumberOfSeedElements" -> RandomInteger[{2, 8}], "ConnectingFunction" -> FilledCurve@*BezierCurve], 12], 6, Background -> Black]
0dwkbztss087q

Simplistic approach

We can just generate a large enough set of mandalas and pick the ones we like.

More precisely we have the following steps:

  1. We generate, say, 200 random mandalas using BlockRandom and keeping track of the random seeds
    1. The mandalas are generated with rotational symmetry order 2 and filled Bezier curve connections.
  2. We pick mandalas that look, more or less, like Bethlehem Stars
  3. Add picked mandalas to the results list
  4. If too few mandalas are in the results list go to 1.

Here are some mandalas generated with those steps:

lsStarReferenceSeeds = DeleteDuplicates@{697734, 227488491, 296515155601, 328716690761, 25979673846, 48784395076, 61082107304, 63772596796, 128581744446, 194807926867, 254647184786, 271909611066, 296515155601, 575775702222, 595562118302, 663386458123, 664847685618, 680328164429, 859482663706};
Multicolumn[
  Table[BlockRandom[ResourceFunction["RandomMandala"]["RotationalSymmetryOrder" -> 2, "NumberOfSeedElements" -> Automatic, "ConnectingFunction" -> FilledCurve@*BezierCurve, ColorFunction -> (White &), Background -> Black], RandomSeeding -> rs], {rs, lsStarReferenceSeeds}] /. GrayLevel[0.25`] -> White, 6, Appearance -> "Horizontal", Background -> Black]
1aedatd1zb3fh

Remark: The plot above looks prettier in notebook converted with the resource function DarkMode.

Elaborated approach

Assume that we want to automate the simplistic approach described in the previous section.

One way to automate is to create a Machine Learning (ML) classifier that is capable of discerning which RandomMandala objects look like Bethlehem Star target images and which do not. With such a classifier we can write a function BethlehemMandala that applies the classifier on multiple results from RandomMandala and returns those mandalas that the classifier says are good.

Here are the steps of building the proposed classifier:

  • Generate a large enough Random Mandala Images Set (RMIS)
  • Create a feature extractor from a subset of RMIS
  • Assign features to all of RMIS
  • Make a recommender with the RMIS features and other image data (like pixel values)
  • Apply the RMIS recommender over the target Bethlehem Star images and determine and examine image sets that are:
    • the best recommendations
    • the worse recommendations
  • With the best and worse recommendations sets compose training data for classifier making
  • Train a classifier
  • Examine classifier application to (filtering of) random mandala images (both in RMIS and not in RMIS)
  • If the results are not satisfactory redo some or all of the steps above

Remark: If the results are not satisfactory we should consider using the obtained classifier at the data generation phase. (This is not done in this document/notebook.)

Remark: The elaborated approach outline and flow chart have general applicability, not just for generation of random images of a certain type.

Flow chart

Here is a flow chart that corresponds to the outline above:

09agsmbmjhhv4

A few observations for the flow chart follow:

  • The flow chart has a feature extraction block that shows that the feature extraction can be done in several ways.
    • The application of LSA is a type of feature extraction which this document/notebook uses.
  • If the results are not good enough the flow chart shows that the classifier can be used at the data generation phase.
  • If the results are not good enough there are several alternatives to redo or tune the ML algorithms.
    • Changing or tuning the recommender implies training a new classifier.
    • Changing or tuning the feature extraction implies making a new recommender and a new classifier.

Data generation and preparation

In this section we generate random mandala graphics, transform them into images and corresponding vectors. Those image-vectors can be used to apply dimension reduction algorithms. (Other feature extraction algorithms can be applied over the images.)

Generated data

Generate large number of mandalas:

k = 20000;
knownSeedsQ = False;
SeedRandom[343];
lsRSeeds = Union@RandomInteger[{1, 10^9}, k];
AbsoluteTiming[
  aMandalas = 
    If[TrueQ@knownSeedsQ, 
     Association@Table[rs -> BlockRandom[ResourceFunction["RandomMandala"]["RotationalSymmetryOrder" -> 2, "NumberOfSeedElements" -> Automatic, "ConnectingFunction" -> FilledCurve@*BezierCurve], RandomSeeding -> rs], {rs, lsRSeeds}], 
    (*ELSE*) 
     Association@Table[i -> ResourceFunction["RandomMandala"]["RotationalSymmetryOrder" -> 2, "NumberOfSeedElements" -> Automatic, "ConnectingFunction" -> FilledCurve@*BezierCurve], {i, 1, k}] 
    ]; 
 ]

(*{18.7549, Null}*)

Check the number of mandalas generated:

Length[aMandalas]

(*20000*)

Show a sample of the generated mandalas:

Magnify[Multicolumn[MandalaToWhiterImage /@ RandomSample[Values@aMandalas, 40], 10, Background -> Black], 0.7]
1gpblane63eo9

Data preparation

Convert the mandala graphics into images using appropriately large (or appropriately small) image sizes:

AbsoluteTiming[
  aMImages = ParallelMap[ImageResize[#, {120, 120}] &, aMandalas]; 
 ]

(*{248.202, Null}*)

Flatten each of the images into vectors:

AbsoluteTiming[
  aMImageVecs = ParallelMap[Flatten[ImageData[Binarize@ColorNegate@ColorConvert[#, "Grayscale"]]] &, aMImages]; 
 ]

(*{16.0125, Null}*)

Remark: Below those vectors are called image-vectors.

Feature extraction

In this section we use the software monad LSAMon, [AA1, AAp1], to do dimension reduction over a subset of random mandala images.

Remark: Other feature extraction methods can be used through the built-in functions FeatureExtraction and FeatureExtract.

Dimension reduction

Create an LSAMon object and extract image topics using Singular Value Decomposition (SVD) or Independent Component Analysis (ICA), [AAr2]:

SeedRandom[893];
AbsoluteTiming[
  lsaObj = 
    LSAMonUnit[]⟹
     LSAMonSetDocumentTermMatrix[SparseArray[Values@RandomSample[aMImageVecs, UpTo[2000]]]]⟹
     LSAMonApplyTermWeightFunctions["None", "None", "Cosine"]⟹
     LSAMonExtractTopics["NumberOfTopics" -> 40, Method -> "ICA", "MaxSteps" -> 240, "MinNumberOfDocumentsPerTerm" -> 0]⟹
     LSAMonNormalizeMatrixProduct[Normalized -> Left]; 
 ]

(*{16.1871, Null}*)

Show the importance coefficients of the topics (if SVD was used the plot would show the singular values):

ListPlot[Norm /@ SparseArray[lsaObj⟹LSAMonTakeH], Filling -> Axis, PlotRange -> All, PlotTheme -> "Scientific"]
1sy1zsgpxysof

Show the interpretation of the extracted image topics:

lsaObj⟹
   LSAMonNormalizeMatrixProduct[Normalized -> Right]⟹
   LSAMonEchoFunctionContext[ImageAdjust[Image[Partition[#, ImageDimensions[aMImages[[1]]][[1]]]]] & /@ SparseArray[#H] &];
16h8a7jwknnkt

Approximation

Pick a test image that is a mandala image or a target image and pre-process it:

If[True, 
   ind = RandomChoice[Range[Length[Values[aMImages]]]]; 
   imgTest = MandalaToWhiterImage@aMandalas[[ind]]; 
   matImageTest = ToSSparseMatrix[SparseArray@List@ImageToVector[imgTest, ImageDimensions[aMImages[[1]]]], "RowNames" -> Automatic, "ColumnNames" -> Automatic], 
  (*ELSE*) 
   imgTest = Binarize[imgStar2, 0.5]; 
   matImageTest = ToSSparseMatrix[SparseArray@List@ImageToVector[imgTest, ImageDimensions[aMImages[[1]]]], "RowNames" -> Automatic, "ColumnNames" -> Automatic] 
  ];
imgTest
0vlq50ryrw0hl

Find the representation of the test image with the chosen feature extractor (LSAMon object here):

matReprsentation = lsaObj⟹LSAMonRepresentByTopics[matImageTest]⟹LSAMonTakeValue;
lsCoeff = Normal@SparseArray[matReprsentation[[1, All]]];
ListPlot[lsCoeff, Filling -> Axis, PlotRange -> All]
1u57b208thtfz

Show the interpretation of the found representation:

H = SparseArray[lsaObj⟹LSAMonNormalizeMatrixProduct[Normalized -> Right]⟹LSAMonTakeH];
vecReprsentation = lsCoeff . H;
ImageAdjust@Image[Rescale[Partition[vecReprsentation, ImageDimensions[aMImages[[1]]][[1]]]]]
1m7r3b5bx32ow

Recommendations

In this section we utilize the software monad SMRMon, [AAp3], to create a recommender for the random mandala images.

Remark: Instead of the Sparse Matrix Recommender (SMR) object the built-in function Nearest can be used.

Create SSparseMatrix object for all image-vectors:

matImages = ToSSparseMatrix[SparseArray[Values@aMImageVecs], "RowNames" -> Automatic, "ColumnNames" -> Automatic]
029x975bs3q7w

Normalize the rows of the image-vectors matrix:

AbsoluteTiming[
  matPixel = WeightTermsOfSSparseMatrix[matImages, "None", "None", "Cosine"] 
 ]
1k9xucwektmhh

Get the LSA topics matrix:

matH = (lsaObj⟹LSAMonNormalizeMatrixProduct[Normalized -> Right]⟹LSAMonTakeH)
05zsn0o1jyqj6

Find the image topics representation for each image-vector (assuming matH was computed with SVD or ICA):

AbsoluteTiming[
  matTopic = matPixel . Transpose[matH] 
 ]
028u1jz1hgzx9

Here we create a recommender based on the images data (pixels) and extracted image topics (or other image features):

smrObj = 
   SMRMonUnit[]⟹
    SMRMonCreate[<|"Pixel" -> matPixel, "Topic" -> matTopic|>]⟹
    SMRMonApplyNormalizationFunction["Cosine"]⟹
    SMRMonSetTagTypeWeights[<|"Pixel" -> 0.2, "Topic" -> 1|>];

Remark: Note the weights assigned to the pixels and the topics in the recommender object above. Those weights were derived by examining the recommendations results shown below.

Here is the image we want to find most similar mandala images to – the target image:

imgTarget = Binarize[imgStar2, 0.5]
1qdmarfxa5i78

Here is the profile of the target image:

aProf = MakeSMRProfile[lsaObj, imgTarget, ImageDimensions[aMImages[[1]]]];
TakeLargest[aProf, 6]

(*<|"10032-10009-4392" -> 0.298371, "3906-10506-10495" -> 0.240086, "10027-10014-4387" -> 0.156797, "8342-8339-6062" -> 0.133822, "3182-3179-11222" -> 0.131565, "8470-8451-5829" -> 0.128844|>*)

Using the target image profile here we compute the recommendation scores for all mandala images of the recommender:

aRecs = 
   smrObj⟹
    SMRMonRecommendByProfile[aProf, All]⟹
    SMRMonTakeValue;

Here is a plot of the similarity scores:

Row[{ResourceFunction["RecordsSummary"][Values[aRecs]], ListPlot[Values[aRecs], ImageSize -> Medium, PlotRange -> All, PlotTheme -> "Detailed", PlotLabel -> "Similarity scores"]}]
1kdiisj4jg4ut

Here are the closest (nearest neighbor) mandala images:

Multicolumn[Values[ImageAdjust@*ColorNegate /@ aMImages[[ToExpression /@ Take[Keys[aRecs], 48]]]], 12, Background -> Black]
096uazw8izidy

Here are the most distant mandala images:

Multicolumn[Values[ImageAdjust@*ColorNegate /@ aMImages[[ToExpression /@ Take[Keys[aRecs], -48]]]], 12, Background -> Black]
0zb7hf24twij4

Classifier creation and utilization

In this section we:

  • Prepare classifier data
  • Build and examine a classifier using the software monad ClCon, [AA2, AAp2], using appropriate training, testing, and validation data ratios
  • Build a classifier utilizing all training data
  • Generate Bethlehem Star mandalas by filtering mandala candidates with the classifier

As it was mentioned above we prepare the data to build classifiers with by:

  • Selecting top, highest scores recommendations and labeling them with True
  • Selecting bad, low score recommendations and labeling them with False
AbsoluteTiming[
  Block[{
    lsBest = Values@aMandalas[[ToExpression /@ Take[Keys[aRecs], 120]]], 
    lsWorse = Values@aMandalas[[ToExpression /@ Join[Take[Keys[aRecs], -200], RandomSample[Take[Keys[aRecs], {3000, -200}], 200]]]]}, 
   lsTrainingData = 
     Join[
      Map[MandalaToWhiterImage[#, ImageDimensions@aMImages[[1]]] -> True &, lsBest], 
      Map[MandalaToWhiterImage[#, ImageDimensions@aMImages[[1]]] -> False &, lsWorse] 
     ]; 
  ] 
 ]

(*{27.9127, Null}*)

Using ClCon train a classifier and show its performance measures:

clObj = 
   ClConUnit[lsTrainingData]⟹
    ClConSplitData[0.75, 0.2]⟹
    ClConMakeClassifier["NearestNeighbors"]⟹
    ClConClassifierMeasurements⟹
    ClConEchoValue⟹
    ClConClassifierMeasurements["ConfusionMatrixPlot"]⟹
    ClConEchoValue;
0jkfza6x72kb5
03uf3deiz0hsd

Remark: We can re-run the ClCon workflow above several times until we obtain a classifier we want to use.

Train a classifier with all prepared data:

clObj2 = 
   ClConUnit[lsTrainingData]⟹
    ClConSplitData[1, 0.2]⟹
    ClConMakeClassifier["NearestNeighbors"];

Get the classifier function from ClCon object:

cfBStar = clObj2⟹ClConTakeClassifier
0awjjib00ihgg

Here we generate Bethlehem Star mandalas using the classifier trained above:

SeedRandom[2020];
Multicolumn[MandalaToWhiterImage /@ BethlehemMandala[12, cfBStar, 0.87], 6, Background -> Black]
0r37g633mpq0y

Generate Bethlehem Star mandala images utilizing the classifier (with a specified classifier probabilities threshold):

SeedRandom[32];
KeyMap[MandalaToWhiterImage, BethlehemMandala[12, cfBStar, 0.87, "Probabilities" -> True]]
0osesxm4gdvvf

Show unfiltered Bethlehem Star mandala candidates:

SeedRandom[32];
KeyMap[MandalaToWhiterImage, BethlehemMandala[12, cfBStar, 0, "Probabilities" -> True]]
0rr12n6savl9z

Remark: Examine the probabilities in the image-probability associations above – they show that the classifier is “working.“

Here is another set generated Bethlehem Star mandalas using rotational symmetry order 4:

SeedRandom[777];
KeyMap[MandalaToWhiterImage, BethlehemMandala[12, cfBStar, 0.8, "RotationalSymmetryOrder" -> 4, "Probabilities" -> True]]
0rgzjquk4amz4

Remark: Note that although a higher rotational symmetry order is used the highly scored results still seem relevant – they have the features of the target Bethlehem Star images.

References

[AA1] Anton Antonov, “A monad for Latent Semantic Analysis workflows”, (2019), MathematicaForPrediction at WordPress.

[AA2] Anton Antonov, “A monad for classification workflows”, (2018)), MathematicaForPrediction at WordPress.

[MSE1] “Plotting the Star of Bethlehem”, (2020),Mathematica Stack Exchange, question 236499,

[Wk1] Wikipedia entry, Star of Bethlehem.

Packages

[AAr1] Anton Antonov, RandomMandala, (2019), Wolfram Function Repository.

[AAr2] Anton Antonov, IdependentComponentAnalysis, (2019), Wolfram Function Repository.

[AAr3] Anton Antonov, “Simplified Machine Learning Workflows” book, (2019), GitHub/antononcube.

[AAp1] Anton Antonov, Monadic Latent Semantic Analysis Mathematica package, (2017), MathematicaForPrediction at GitHub/antononcube.

[AAp2] Anton Antonov, Monadic contextual classification Mathematica package, (2017), MathematicaForPrediction at GitHub/antononcube.

[AAp3] Anton Antonov, Monadic Sparse Matrix Recommender Mathematica package, (2018), MathematicaForPrediction at GitHub/antononcube.

Code definitions

urlPart = "https://raw.githubusercontent.com/antononcube/MathematicaForPrediction/master/MonadicProgramming/";
Get[urlPart <> "MonadicLatentSemanticAnalysis.m"];
Get[urlPart <> "MonadicSparseMatrixRecommender.m"];
Get[urlPart <> "/MonadicContextualClassification.m"];
Clear[MandalaToImage, MandalaToWhiterImage];
MandalaToImage[gr_Graphics, imgSize_ : {120, 120}] := ColorNegate@ImageResize[gr, imgSize];
MandalaToWhiterImage[gr_Graphics, imgSize_ : {120, 120}] := ColorNegate@ImageResize[gr /. GrayLevel[0.25`] -> Black, imgSize];
Clear[ImageToVector];
ImageToVector[img_Image] := Flatten[ImageData[ColorConvert[img, "Grayscale"]]];
ImageToVector[img_Image, imgSize_] := Flatten[ImageData[ColorConvert[ImageResize[img, imgSize], "Grayscale"]]];
ImageToVector[___] := $Failed;
Clear[MakeSMRProfile];
MakeSMRProfile[lsaObj_LSAMon, gr_Graphics, imgSize_] := MakeSMRProfile[lsaObj, {gr}, imgSize];
MakeSMRProfile[lsaObj_LSAMon, lsGrs : {_Graphics}, imgSize_] := MakeSMRProfile[lsaObj, MandalaToWhiterImage[#, imgSize] & /@ lsGrs, imgSize]
MakeSMRProfile[lsaObj_LSAMon, img_Image, imgSize_] := MakeSMRProfile[lsaObj, {img}, imgSize];
MakeSMRProfile[lsaObj_LSAMon, lsImgs : {_Image ..}, imgSize_] := 
   Block[{lsImgVecs, matTest, aProfPixel, aProfTopic}, 
    lsImgVecs = ImageToVector[#, imgSize] & /@ lsImgs; 
    matTest = ToSSparseMatrix[SparseArray[lsImgVecs], "RowNames" -> Automatic, "ColumnNames" -> Automatic]; 
    aProfPixel = ColumnSumsAssociation[lsaObj⟹LSAMonRepresentByTerms[matTest]⟹LSAMonTakeValue]; 
    aProfTopic = ColumnSumsAssociation[lsaObj⟹LSAMonRepresentByTopics[matTest]⟹LSAMonTakeValue]; 
    aProfPixel = Select[aProfPixel, # > 0 &]; 
    aProfTopic = Select[aProfTopic, # > 0 &]; 
    Join[aProfPixel, aProfTopic] 
   ];
MakeSMRProfile[___] := $Failed;
Clear[BethlehemMandalaCandiate];
BethlehemMandalaCandiate[opts : OptionsPattern[]] := ResourceFunction["RandomMandala"][opts, "RotationalSymmetryOrder" -> 2, "NumberOfSeedElements" -> Automatic, "ConnectingFunction" -> FilledCurve@*BezierCurve];
Clear[BethlehemMandala];
Options[BethlehemMandala] = Join[{ImageSize -> {120, 120}, "Probabilities" -> False}, Options[ResourceFunction["RandomMandala"]]];
BethlehemMandala[n_Integer, cf_ClassifierFunction, opts : OptionsPattern[]] := BethlehemMandala[n, cf, 0.87, opts];
BethlehemMandala[n_Integer, cf_ClassifierFunction, threshold_?NumericQ, opts : OptionsPattern[]] := 
   Block[{imgSize, probsQ, res, resNew, aResScores = <||>, aResScoresNew = <||>}, 
     
     imgSize = OptionValue[BethlehemMandala, ImageSize]; 
     probsQ = TrueQ[OptionValue[BethlehemMandala, "Probabilities"]]; 
     
     res = {}; 
     While[Length[res] < n, 
      resNew = Table[BethlehemMandalaCandiate[FilterRules[{opts}, Options[ResourceFunction["RandomMandala"]]]], 2*(n - Length[res])]; 
      aResScoresNew = Association[# -> cf[MandalaToImage[#, imgSize], "Probabilities"][True] & /@ resNew]; 
      aResScoresNew = Select[aResScoresNew, # >= threshold &]; 
      aResScores = Join[aResScores, aResScoresNew]; 
      res = Keys[aResScores] 
     ]; 
     
     aResScores = TakeLargest[ReverseSort[aResScores], UpTo[n]]; 
     If[probsQ, aResScores, Keys[aResScores]] 
    ] /; n > 0;
BethlehemMandala[___] := $Failed

A monad for Latent Semantic Analysis workflows

Introduction

In this document we describe the design and implementation of a (software programming) monad, [Wk1], for Latent Semantic Analysis workflows specification and execution. The design and implementation are done with Mathematica / Wolfram Language (WL).

What is Latent Semantic Analysis (LSA)? : A statistical method (or a technique) for finding relationships in natural language texts that is based on the so called Distributional hypothesis, [Wk2, Wk3]. (The Distributional hypothesis can be simply stated as “linguistic items with similar distributions have similar meanings”; for an insightful philosophical and scientific discussion see [MS1].) LSA can be seen as the application of Dimensionality reduction techniques over matrices derived with the Vector space model.

The goal of the monad design is to make the specification of LSA workflows (relatively) easy and straightforward by following a certain main scenario and specifying variations over that scenario.

The monad is named LSAMon and it is based on the State monad package “StateMonadCodeGenerator.m”, [AAp1, AA1], the document-term matrix making package “DocumentTermMatrixConstruction.m”, [AAp4, AA2], the Non-Negative Matrix Factorization (NNMF) package “NonNegativeMatrixFactorization.m”, [AAp5, AA2], and the package “SSparseMatrix.m”, [AAp2, AA5], that provides matrix objects with named rows and columns.

The data for this document is obtained from WL’s repository and it is manipulated into a certain ready-to-utilize form (and uploaded to GitHub.)

The monadic programming design is used as a Software Design Pattern. The LSAMon monad can be also seen as a Domain Specific Language (DSL) for the specification and programming of machine learning classification workflows.

Here is an example of using the LSAMon monad over a collection of documents that consists of 233 US state of union speeches.

LSAMon-Introduction-pipeline
LSAMon-Introduction-pipeline
LSAMon-Introduction-pipeline-echos
LSAMon-Introduction-pipeline-echos

The table above is produced with the package “MonadicTracing.m”, [AAp2, AA1], and some of the explanations below also utilize that package.

As it was mentioned above the monad LSAMon can be seen as a DSL. Because of this the monad pipelines made with LSAMon are sometimes called “specifications”.

Remark: In this document with “term” we mean “a word, a word stem, or other type of token.”

Remark: LSA and Latent Semantic Indexing (LSI) are considered more or less to be synonyms. I think that “latent semantic analysis” sounds more universal and that “latent semantic indexing” as a name refers to a specific Information Retrieval technique. Below we refer to “LSI functions” like “IDF” and “TF-IDF” that are applied within the generic LSA workflow.

Contents description

The document has the following structure.

  • The sections “Package load” and “Data load” obtain the needed code and data.
  • The sections “Design consideration” and “Monad design” provide motivation and design decisions rationale.
  • The sections “LSAMon overview”, “Monad elements”, and “The utilization of SSparseMatrix objects” provide technical descriptions needed to utilize the LSAMon monad .
    • (Using a fair amount of examples.)
  • The section “Unit tests” describes the tests used in the development of the LSAMon monad.
    • (The random pipelines unit tests are especially interesting.)
  • The section “Future plans” outlines future directions of development.
  • The section “Implementation notes” just says that LSAMon’s development process and this document follow the ones of the classifications workflows monad ClCon, [AA6].

Remark: One can read only the sections “Introduction”, “Design consideration”, “Monad design”, and “LSAMon overview”. That set of sections provide a fairly good, programming language agnostic exposition of the substance and novel ideas of this document.

Package load

The following commands load the packages [AAp1–AAp7]:

Import["https://raw.githubusercontent.com/antononcube/MathematicaForPrediction/master/MonadicProgramming/MonadicLatentSemanticAnalysis.m"]
Import["https://raw.githubusercontent.com/antononcube/MathematicaForPrediction/master/MonadicProgramming/MonadicTracing.m"]

Data load

In this section we load data that is used in the rest of the document. The text data was obtained through WL’s repository, transformed in a certain more convenient form, and uploaded to GitHub.

The text summarization and plots are done through LSAMon, which in turn uses the function RecordsSummary from the package “MathematicaForPredictionUtilities.m”, [AAp7].

Hamlet

textHamlet = 
  ToString /@ 
   Flatten[Import["https://raw.githubusercontent.com/antononcube/MathematicaVsR/master/Data/MathematicaVsR-Data-Hamlet.csv"]];

TakeLargestBy[
 Tally[DeleteStopwords[ToLowerCase[Flatten[TextWords /@ textHamlet]]]], #[[2]] &, 20]

(* {{"ham", 358}, {"lord", 225}, {"king", 196}, {"o", 124}, {"queen", 120}, 
    {"shall", 114}, {"good", 109}, {"hor", 109}, {"come",  107}, {"hamlet", 107}, 
    {"thou", 105}, {"let", 96}, {"thy", 86}, {"pol", 86}, {"like", 81}, {"sir", 75}, 
    {"'t", 75}, {"know", 74}, {"enter", 73}, {"th", 72}} *)

LSAMonUnit[textHamlet]⟹LSAMonMakeDocumentTermMatrix⟹LSAMonEchoDocumentTermMatrixStatistics;
LSAMon-Data-Load-Hamlet-echo
LSAMon-Data-Load-Hamlet-echo

USA state of union speeches

url = "https://github.com/antononcube/MathematicaVsR/blob/master/Data/MathematicaVsR-Data-StateOfUnionSpeeches.JSON.zip?raw=true";
str = Import[url, "String"];
filename = First@Import[StringToStream[str], "ZIP"];
aStateOfUnionSpeeches = Association@ImportString[Import[StringToStream[str], {"ZIP", filename, "String"}], "JSON"];

lsaObj = 
LSAMonUnit[aStateOfUnionSpeeches]⟹
LSAMonMakeDocumentTermMatrix⟹
LSAMonEchoDocumentTermMatrixStatistics["LogBase" -> 10];
LSAMon-Data-Load-StateOfUnionSpeeches-echo
LSAMon-Data-Load-StateOfUnionSpeeches-echo
TakeLargest[ColumnSumsAssociation[lsaObj⟹LSAMonTakeDocumentTermMatrix], 12]

(* <|"government" -> 7106, "states" -> 6502, "congress" -> 5023, 
     "united" -> 4847, "people" -> 4103, "year" -> 4022, 
     "country" -> 3469, "great" -> 3276, "public" -> 3094, "new" -> 3022, 
     "000" -> 2960, "time" -> 2922|> *)

Stop words

In some of the examples below we want to explicitly specify the stop words. Here are stop words derived using the built-in functions DictionaryLookup and DeleteStopwords.

stopWords = Complement[DictionaryLookup["*"], DeleteStopwords[DictionaryLookup["*"]]];

Short[stopWords]

(* {"a", "about", "above", "across", "add-on", "after", "again", <<290>>, 
   "you'll", "your", "you're", "yours", "yourself", "yourselves", "you've" } *)
    

Design considerations

The steps of the main LSA workflow addressed in this document follow.

  1. Get a collection of documents with associated ID’s.
  2. Create a document-term matrix.
    1. Here we apply the Bag-or-words model and Vector space model.
      1. The sequential order of the words is ignored and each document is represented as a point in a multi-dimensional vector space.
      2. That vector space axes correspond to the unique words found in the whole document collection.
    2. Consider the application of stemming rules.
    3. Consider the removal of stop words.
  3. Apply matrix-entries weighting functions.
    1. Those functions come from LSI.
    2. Functions like “IDF”, “TF-IDF”, “GFIDF”.
  4. Extract topics.
    1. One possible statistical way of doing this is with Dimensionality reduction.
    2. We consider using Singular Value Decomposition (SVD) and Non-Negative Matrix Factorization (NNMF).
  5. Make and display the topics table.
  6. Extract and display a statistical thesaurus of selected words.
  7. Map search queries or unseen documents over the extracted topics.
  8. Find the most important documents in the document collection. (Optional.)

The following flow-chart corresponds to the list of steps above.

LSA-worflows
LSA-worflows

In order to address:

  • the introduction of new elements in LSA workflows,
  • workflows elements variability, and
  • workflows iterative changes and refining,

it is beneficial to have a DSL for LSA workflows. We choose to make such a DSL through a functional programming monad, [Wk1, AA1].

Here is a quote from [Wk1] that fairly well describes why we choose to make a classification workflow monad and hints on the desired properties of such a monad.

[…] The monad represents computations with a sequential structure: a monad defines what it means to chain operations together. This enables the programmer to build pipelines that process data in a series of steps (i.e. a series of actions applied to the data), in which each action is decorated with the additional processing rules provided by the monad. […]

Monads allow a programming style where programs are written by putting together highly composable parts, combining in flexible ways the possible actions that can work on a particular type of data. […]

Remark: Note that quote from [Wk1] refers to chained monadic operations as “pipelines”. We use the terms “monad pipeline” and “pipeline” below.

Monad design

The monad we consider is designed to speed-up the programming of LSA workflows outlined in the previous section. The monad is named LSAMon for “Latent Semantic Analysis** Mon**ad”.

We want to be able to construct monad pipelines of the general form:

LSAMon-Monad-Design-formula-1
LSAMon-Monad-Design-formula-1

LSAMon is based on the State monad, [Wk1, AA1], so the monad pipeline form (1) has the following more specific form:

LSAMon-Monad-Design-formula-2
LSAMon-Monad-Design-formula-2

This means that some monad operations will not just change the pipeline value but they will also change the pipeline context.

In the monad pipelines of LSAMon we store different objects in the contexts for at least one of the following two reasons.

  1. The object will be needed later on in the pipeline, or
  2. The object is (relatively) hard to compute.

Such objects are document-term matrix, Dimensionality reduction factors and the related topics.

Let us list the desired properties of the monad.

  • Rapid specification of non-trivial LSA workflows.
  • The text data supplied to the monad can be: (i) a list of strings, or (ii) an association with string values.
  • The monad uses the Linear vector space model .
  • The document-term frequency matrix can be created after removing stop words and/or word stemming.
  • It is easy to specify and apply different LSI weight functions. (Like “IDF” or “GFIDF”.)
  • The monad can do dimension reduction with SVD and NNMF and corresponding matrix factors are retrievable with monad functions.
  • Documents (or query strings) external to the monad are easily mapped into monad’s Linear vector space of terms and Linear vector space of topics.
  • The monad allows of cursory examination and summarization of the data.
  • The pipeline values can be of different types. (Most monad functions modify the pipeline value; some modify the context; some just echo results.)
  • It is easy to obtain the pipeline value, context, and different context objects for manipulation outside of the monad.
  • It is easy to tabulate extracted topics and related statistical thesauri.

The LSAMon components and their interactions are fairly simple.

The main LSAMon operations implicitly put in the context or utilize from the context the following objects:

  • document-term matrix,
  • the factors obtained by matrix factorization algorithms,
  • LSI weight functions specifications,
  • extracted topics.

Note the that the monadic set of types of LSAMon pipeline values is fairly heterogenous and certain awareness of “the current pipeline value” is assumed when composing LSAMon pipelines.

Obviously, we can put in the context any object through the generic operations of the State monad of the package “StateMonadGenerator.m”, [AAp1].

LSAMon overview

When using a monad we lift certain data into the “monad space”, using monad’s operations we navigate computations in that space, and at some point we take results from it.

With the approach taken in this document the “lifting” into the LSAMon monad is done with the function LSAMonUnit. Results from the monad can be obtained with the functions LSAMonTakeValue, LSAMonContext, or with the other LSAMon functions with the prefix “LSAMonTake” (see below.)

Here is a corresponding diagram of a generic computation with the LSAMon monad:

LSAMon-pipeline
LSAMon-pipeline

Remark: It is a good idea to compare the diagram with formulas (1) and (2).

Let us examine a concrete LSAMon pipeline that corresponds to the diagram above. In the following table each pipeline operation is combined together with a short explanation and the context keys after its execution.

Here is the output of the pipeline:

The LSAMon functions are separated into four groups:

  • operations,
  • setters and droppers,
  • takers,
  • State Monad generic functions.

Monad functions interaction with the pipeline value and context

An overview of the those functions is given in the tables in next two sub-sections. The next section, “Monad elements”, gives details and examples for the usage of the LSAMon operations.

LSAMon-Overview-operations-context-interactions-table
LSAMon-Overview-operations-context-interactions-table
LSAMon-Overview-setters-droppers-takers-context-interactions-table
LSAMon-Overview-setters-droppers-takers-context-interactions-table

State monad functions

Here are the LSAMon State Monad functions (generated using the prefix “LSAMon”, [AAp1, AA1].)

LSAMon-Overview-StMon-usage-descriptions-table
LSAMon-Overview-StMon-usage-descriptions-table

Main monad functions

Here are the usage descriptions of the main (not monad-supportive) LSAMon functions, which are explained in detail in the next section.

LSAMon-Overview-operations-usage-descriptions-table
LSAMon-Overview-operations-usage-descriptions-table

Monad elements

In this section we show that LSAMon has all of the properties listed in the previous section.

The monad head

The monad head is LSAMon. Anything wrapped in LSAMon can serve as monad’s pipeline value. It is better though to use the constructor LSAMonUnit. (Which adheres to the definition in [Wk1].)

LSAMon[textHamlet, <||>]⟹LSAMonMakeDocumentTermMatrix[Automatic, Automatic]⟹LSAMonEchoFunctionContext[Short];

Lifting data to the monad

The function lifting the data into the monad LSAMon is LSAMonUnit.

The lifting to the monad marks the beginning of the monadic pipeline. It can be done with data or without data. Examples follow.

LSAMonUnit[textHamlet]⟹LSAMonMakeDocumentTermMatrix⟹LSAMonTakeDocumentTermMatrix

LSAMonUnit[]⟹LSAMonSetDocuments[textHamlet]⟹LSAMonMakeDocumentTermMatrix⟹LSAMonTakeDocumentTermMatrix

(See the sub-section “Setters, droppers, and takers” for more details of setting and taking values in LSAMon contexts.)

Currently the monad can deal with data in the following forms:

  • vectors of strings,
  • associations with string values.

Generally, WL makes it easy to extract columns datasets order to obtain vectors or matrices, so datasets are not currently supported in LSAMon.

Making of the document-term matrix

As it was mentioned above with “term” we mean “a word or a stemmed word”. Here is are examples of stemmed words.

WordData[#, "PorterStem"] & /@ {"consequential", "constitution", "forcing", ""}

The fundamental model of LSAMon is the so called Vector space model (or the closely related Bag-of-words model.) The document-term matrix is a linear vector space representation of the documents collection. That representation is further used in LSAMon to find topics and statistical thesauri.

Here is an example of ad hoc construction of a document-term matrix using a couple of paragraphs from “Hamlet”.

inds = {10, 19};
aTempText = AssociationThread[inds, textHamlet[[inds]]]

MatrixForm @ CrossTabulate[Flatten[KeyValueMap[Thread[{#1, #2}] &, TextWords /@ ToLowerCase[aTempText]], 1]]

When we construct the document-term matrix we (often) want to stem the words and (almost always) want to remove stop words. LSAMon’s function LSAMonMakeDocumentTermMatrix makes the document-term matrix and takes specifications for stemming and stop words.

lsaObj =
  LSAMonUnit[textHamlet]⟹
   LSAMonMakeDocumentTermMatrix["StemmingRules" -> Automatic, "StopWords" -> Automatic]⟹
   LSAMonEchoFunctionContext[ MatrixPlot[#documentTermMatrix] &]⟹
   LSAMonEchoFunctionContext[TakeLargest[ColumnSumsAssociation[#documentTermMatrix], 12] &];

We can retrieve the stop words used in a monad with the function LSAMonTakeStopWords.

Short[lsaObj⟹LSAMonTakeStopWords]

We can retrieve the stemming rules used in a monad with the function LSAMonTakeStemmingRules.

Short[lsaObj⟹LSAMonTakeStemmingRules]

The specification Automatic for stemming rules uses WordData[#,"PorterStem"]&.

Instead of the options style signature we can use positional signature.

  • Options style: LSAMonMakeDocumentTermMatrix["StemmingRules" -> {}, "StopWords" -> Automatic] .
  • Positional style: LSAMonMakeDocumentTermMatrix[{}, Automatic] .

LSI weight functions

After making the document-term matrix we will most likely apply LSI weight functions, [Wk2], like “GFIDF” and “TF-IDF”. (This follows the “standard” approach used in search engines for calculating weights for document-term matrices; see [MB1].)

Frequency matrix

We use the following definition of the frequency document-term matrix F.

Each entry fij of the matrix F is the number of occurrences of the term j in the document i.

Weights

Each entry of the weighted document-term matrix M derived from the frequency document-term matrix F is expressed with the formula

where gj – global term weight; lij – local term weight; di – normalization weight.

Various formulas exist for these weights and one of the challenges is to find the right combination of them when using different document collections.

Here is a table of weight functions formulas.

LSAMon-LSI-weight-functions-table
LSAMon-LSI-weight-functions-table

Computation specifications

LSAMon function LSAMonApplyTermWeightFunctions delegates the LSI weight functions application to the package “DocumentTermMatrixConstruction.m”, [AAp4].

Here is an example.

lsaHamlet = LSAMonUnit[textHamlet]⟹LSAMonMakeDocumentTermMatrix;
wmat =
  lsaHamlet⟹
   LSAMonApplyTermWeightFunctions["IDF", "TermFrequency", "Cosine"]⟹
   LSAMonTakeWeightedDocumentTermMatrix;

TakeLargest[ColumnSumsAssociation[wmat], 6]

Instead of using the positional signature of LSAMonApplyTermWeightFunctions we can specify the LSI functions using options.

wmat2 =
  lsaHamlet⟹
   LSAMonApplyTermWeightFunctions["GlobalWeightFunction" -> "IDF", "LocalWeightFunction" -> "TermFrequency", "NormalizerFunction" -> "Cosine"]⟹
   LSAMonTakeWeightedDocumentTermMatrix;

TakeLargest[ColumnSumsAssociation[wmat2], 6]

Here we are summaries of the non-zero values of the weighted document-term matrix derived with different combinations of global, local, and normalization weight functions.

Magnify[#, 0.8] &@Multicolumn[Framed /@ #, 6] &@Flatten@
  Table[
   (wmat =
     lsaHamlet⟹
      LSAMonApplyTermWeightFunctions[gw, lw, nf]⟹
      LSAMonTakeWeightedDocumentTermMatrix;
    RecordsSummary[SparseArray[wmat]["NonzeroValues"], 
     List@StringRiffle[{gw, lw, nf}, ", "]]),
   {gw, {"IDF", "GFIDF", "Binary", "None", "ColumnStochastic"}},
   {lw, {"Binary", "Log", "None"}},
   {nf, {"Cosine", "None", "RowStochastic"}}]
AutoCollapse[]
LSAMon-LSI-weight-functions-combinations-application-table
LSAMon-LSI-weight-functions-combinations-application-table

Extracting topics

Streamlining topic extraction is one of the main reasons LSAMon was implemented. The topic extraction correspond to the so called “syntagmatic” relationships between the terms, [MS1].

Theoretical outline

The original weighed document-term matrix M is decomposed into the matrix factors W and H.

M ≈ W.H, W ∈ ℝm × k, H ∈ ℝk × n.

The i-th row of M is expressed with the i-th row of W multiplied by H.

The rows of H are the topics. SVD produces orthogonal topics; NNMF does not.

The i-the document of the collection corresponds to the i-th row W. Finding the Nearest Neighbors (NN’s) of the i-th document using the rows similarity of the matrix W gives document NN’s through topic similarity.

The terms correspond to the columns of H. Finding NN’s based on similarities of H’s columns produces statistical thesaurus entries.

The term groups provided by H’s rows correspond to “syntagmatic” relationships. Using similarities of H’s columns we can produce term clusters that correspond to “paradigmatic” relationships.

Computation specifications

Here is an example using the play “Hamlet” in which we specify additional stop words.

stopWords2 = {"enter", "exit", "[exit", "ham", "hor", "laer", "pol", "oph", "thy", "thee", "act", "scene"};

SeedRandom[2381]
lsaHamlet =
  LSAMonUnit[textHamlet]⟹
   LSAMonMakeDocumentTermMatrix["StemmingRules" -> Automatic, "StopWords" -> Join[stopWords, stopWords2]]⟹
   LSAMonApplyTermWeightFunctions["GlobalWeightFunction" -> "IDF", "LocalWeightFunction" -> "None", "NormalizerFunction" -> "Cosine"]⟹
   LSAMonExtractTopics["NumberOfTopics" -> 12, "MinNumberOfDocumentsPerTerm" -> 10, Method -> "NNMF", "MaxSteps" -> 20]⟹
   LSAMonEchoTopicsTable["NumberOfTableColumns" -> 6, "NumberOfTerms" -> 10];
LSAMon-Extracting-topics-Hamlet-topics-table
LSAMon-Extracting-topics-Hamlet-topics-table

Here is an example using the USA presidents “state of union” speeches.

SeedRandom[7681]
lsaSpeeches =
  LSAMonUnit[aStateOfUnionSpeeches]⟹
   LSAMonMakeDocumentTermMatrix["StemmingRules" -> Automatic,  "StopWords" -> Automatic]⟹
   LSAMonApplyTermWeightFunctions["GlobalWeightFunction" -> "IDF", "LocalWeightFunction" -> "None", "NormalizerFunction" -> "Cosine"]⟹
   LSAMonExtractTopics["NumberOfTopics" -> 36, "MinNumberOfDocumentsPerTerm" -> 40, Method -> "NNMF", "MaxSteps" -> 12]⟹
   LSAMonEchoTopicsTable["NumberOfTableColumns" -> 6, "NumberOfTerms" -> 10];
LSAMon-Extracting-topics-StateOfUnionSpeeches-topics-table
LSAMon-Extracting-topics-StateOfUnionSpeeches-topics-table

Note that in both examples:

  1. stemming is used when creating the document-term matrix,
  2. the default LSI re-weighting functions are used: “IDF”, “None”, “Cosine”,
  3. the dimension reduction algorithm NNMF is used.

Things to keep in mind.

  1. The interpretability provided by NNMF comes at a price.
  2. NNMF is prone to get stuck into local minima, so several topic extractions (and corresponding evaluations) have to be done.
  3. We would get different results with different NNMF runs using the same parameters. (NNMF uses random numbers initialization.)
  4. The NNMF topic vectors are not orthogonal.
  5. SVD is much faster than NNMF, but it topic vectors are hard to interpret.
  6. Generally, the topics derived with SVD are stable, they do not change with different runs with the same parameters.
  7. The SVD topics vectors are orthogonal, which provides for quick to find representations of documents not in the monad’s document collection.

The document-topic matrix W has column names that are automatically derived from the top three terms in each topic.

ColumnNames[lsaHamlet⟹LSAMonTakeW]

(* {"player-plai-welcom", "ro-lord-sir", "laert-king-attend",
    "end-inde-make", "state-room-castl", "daughter-pass-love",
    "hamlet-ghost-father", "father-thou-king",
    "rosencrantz-guildenstern-king", "ophelia-queen-poloniu",
    "answer-sir-mother", "horatio-attend-gentleman"} *)

Of course the row names of H have the same names.

RowNames[lsaHamlet⟹LSAMonTakeH]

(* {"player-plai-welcom", "ro-lord-sir", "laert-king-attend",
    "end-inde-make", "state-room-castl", "daughter-pass-love",
    "hamlet-ghost-father", "father-thou-king",
    "rosencrantz-guildenstern-king", "ophelia-queen-poloniu",
    "answer-sir-mother", "horatio-attend-gentleman"} *)

Extracting statistical thesauri

The statistical thesaurus extraction corresponds to the “paradigmatic” relationships between the terms, [MS1].

Here is an example over the State of Union speeches.

entryWords = {"bank", "war", "economy", "school", "port", "health", "enemy", "nuclear"};

lsaSpeeches⟹
  LSAMonExtractStatisticalThesaurus["Words" -> Map[WordData[#, "PorterStem"] &, entryWords], "NumberOfNearestNeighbors" -> 12]⟹
  LSAMonEchoStatisticalThesaurus;
LSAMon-Extracting-statistical-thesauri-echo
LSAMon-Extracting-statistical-thesauri-echo

In the code above: (i) the options signature style is used, (ii) the statistical thesaurus entry words are stemmed first.

We can also call LSAMonEchoStatisticalThesaurus directly without calling LSAMonExtractStatisticalThesaurus first.

 lsaSpeeches⟹
   LSAMonEchoStatisticalThesaurus["Words" -> Map[WordData[#, "PorterStem"] &, entryWords], "NumberOfNearestNeighbors" -> 12];
LSAMon-Extracting-statistical-thesauri-echo
LSAMon-Extracting-statistical-thesauri-echo

Mapping queries and documents to terms

One of the most natural operations is to find the representation of an arbitrary document (or sentence or a list of words) in monad’s Linear vector space of terms. This is done with the function LSAMonRepresentByTerms.

Here is an example in which a sentence is represented as a one-row matrix (in that space.)

obj =
  lsaHamlet⟹
   LSAMonRepresentByTerms["Hamlet, Prince of Denmark killed the king."]⟹
   LSAMonEchoValue;

Here we display only the non-zero columns of that matrix.

obj⟹
  LSAMonEchoFunctionValue[MatrixForm[Part[#, All, Keys[Select[SSparseMatrix`ColumnSumsAssociation[#], # > 0& ]]]]& ];

Transformation steps

Assume that LSAMonRepresentByTerms is given a list of sentences. Then that function performs the following steps.

1. The sentence is split into a list of words.

2. If monad’s document-term matrix was made by removing stop words the same stop words are removed from the list of words.

3. If monad’s document-term matrix was made by stemming the same stemming rules are applied to the list of words.

4. The LSI global weights and the LSI local weight and normalizer functions are applied to sentence’s contingency matrix.

Equivalent representation

Let us look convince ourselves that documents used in the monad to built the weighted document-term matrix have the same representation as the corresponding rows of that matrix.

Here is an association of documents from monad’s document collection.

inds = {6, 10};
queries = Part[lsaHamlet⟹LSAMonTakeDocuments, inds];
queries
 
(* <|"id.0006" -> "Getrude, Queen of Denmark, mother to Hamlet. Ophelia, daughter to Polonius.", 
     "id.0010" -> "ACT I. Scene I. Elsinore. A platform before the Castle."|> *)

lsaHamlet⟹
  LSAMonRepresentByTerms[queries]⟹
  LSAMonEchoFunctionValue[MatrixForm[Part[#, All, Keys[Select[SSparseMatrix`ColumnSumsAssociation[#], # > 0& ]]]]& ];
LSAMon-Mapping-queries-and-documents-to-topics-query-matrix
LSAMon-Mapping-queries-and-documents-to-topics-query-matrix
lsaHamlet⟹
  LSAMonEchoFunctionContext[MatrixForm[Part[Slot["weightedDocumentTermMatrix"], inds, Keys[Select[SSparseMatrix`ColumnSumsAssociation[Part[Slot["weightedDocumentTermMatrix"], inds, All]], # > 0& ]]]]& ];
LSAMon-Mapping-queries-and-documents-to-topics-context-sub-matrix
LSAMon-Mapping-queries-and-documents-to-topics-context-sub-matrix

Mapping queries and documents to topics

Another natural operation is to find the representation of an arbitrary document (or a list of words) in monad’s Linear vector space of topics. This is done with the function LSAMonRepresentByTopics.

Here is an example.

inds = {6, 10};
queries = Part[lsaHamlet⟹LSAMonTakeDocuments, inds];
Short /@ queries

(* <|"id.0006" -> "Getrude, Queen of Denmark, mother to Hamlet. Ophelia, daughter to Polonius.", 
     "id.0010" -> "ACT I. Scene I. Elsinore. A platform before the Castle."|> *)

lsaHamlet⟹
  LSAMonRepresentByTopics[queries]⟹
  LSAMonEchoFunctionValue[MatrixForm[Part[#, All, Keys[Select[SSparseMatrix`ColumnSumsAssociation[#], # > 0& ]]]]& ];
LSAMon-Mapping-queries-and-documents-to-terms-query-matrix
LSAMon-Mapping-queries-and-documents-to-terms-query-matrix
lsaHamlet⟹
  LSAMonEchoFunctionContext[MatrixForm[Part[Slot["W"], inds, Keys[Select[SSparseMatrix`ColumnSumsAssociation[Part[Slot["W"], inds, All]], # > 0& ]]]]& ];
LSAMon-Mapping-queries-and-documents-to-terms-query-matrix
LSAMon-Mapping-queries-and-documents-to-terms-query-matrix

Theory

In order to clarify what the function LSAMonRepresentByTopics is doing let us go through the formulas it is based on.

The original weighed document-term matrix M is decomposed into the matrix factors W and H.

M ≈ W.H, W ∈ ℝm × k, H ∈ ℝk × n

The i-th row of M is expressed with the i-th row of W multiplied by H.

mi ≈ wi.H.

For a query vector q0 ∈ ℝm we want to find its topics representation vector x ∈ ℝk:

q0 ≈ x.H.

Denote with H( − 1) the inverse or pseudo-inverse matrix of H. We have:

q0.H( − 1) ≈ (x.H).H( − 1) = x.(H.H( − 1)) = x.I,

x ∈ ℝk, H( − 1) ∈ ℝn × k, I ∈ ℝk × k.

In LSAMon for SVD H( − 1) = HT; for NNMF H( − 1) is the pseudo-inverse of H.

The vector x obtained with LSAMonRepresentByTopics.

Tags representation

Sometimes we want to find the topics representation of tags associated with monad’s documents and the tag-document associations are one-to-many. See [AA3].

Let us consider a concrete example – we want to find what topics correspond to the different presidents in the collection of State of Union speeches.

Here we find the document tags (president names in this case.)

tags = StringReplace[
   RowNames[
    lsaSpeeches⟹LSAMonTakeDocumentTermMatrix], 
   RegularExpression[".\\d\\d\\d\\d-\\d\\d-\\d\\d"] -> ""];
Short[tags]

Here is the number of unique tags (president names.)

Length[Union[tags]]
(* 42 *)

Here we compute the tag-topics representation matrix using the function LSAMonRepresentDocumentTagsByTopics.

tagTopicsMat =
 lsaSpeeches⟹
  LSAMonRepresentDocumentTagsByTopics[tags]⟹
  LSAMonTakeValue

Here is a heatmap plot of the tag-topics matrix made with the package “HeatmapPlot.m”, [AAp11].

HeatmapPlot[tagTopicsMat[[All, Ordering@ColumnSums[tagTopicsMat]]], DistanceFunction -> None, ImageSize -> Large]
LSAMon-Tags-representation-heatmap
LSAMon-Tags-representation-heatmap

Finding the most important documents

There are several algorithms we can apply for finding the most important documents in the collection. LSAMon utilizes two types algorithms: (1) graph centrality measures based, and (2) matrix factorization based. With certain graph centrality measures the two algorithms are equivalent. In this sub-section we demonstrate the matrix factorization algorithm (that uses SVD.)

Definition: The most important sentences have the most important words and the most important words are in the most important sentences.

That definition can be used to derive an iterations-based model that can be expressed with SVD or eigenvector finding algorithms, [LE1].

Here we pick an important part of the play “Hamlet”.

focusText = 
  First@Pick[textHamlet, StringMatchQ[textHamlet, ___ ~~ "to be" ~~ __ ~~ "or not to be" ~~ ___, IgnoreCase -> True]];
Short[focusText]

(* "Ham. To be, or not to be- that is the question: Whether 'tis ....y. 
    O, woe is me T' have seen what I have seen, see what I see!" *)

LSAMonUnit[StringSplit[ToLowerCase[focusText], {",", ".", ";", "!", "?"}]]⟹
  LSAMonMakeDocumentTermMatrix["StemmingRules" -> {}, "StopWords" -> Automatic]⟹
  LSAMonApplyTermWeightFunctions⟹
  LSAMonFindMostImportantDocuments[3]⟹
  LSAMonEchoFunctionValue[GridTableForm];
LSAMon-Find-most-important-documents-table
LSAMon-Find-most-important-documents-table

Setters, droppers, and takers

The values from the monad context can be set, obtained, or dropped with the corresponding “setter”, “dropper”, and “taker” functions as summarized in a previous section.

For example:

p = LSAMonUnit[textHamlet]⟹LSAMonMakeDocumentTermMatrix[Automatic, Automatic];

p⟹LSAMonTakeMatrix

If other values are put in the context they can be obtained through the (generic) function LSAMonTakeContext, [AAp1]:

Short@(p⟹QRMonTakeContext)["documents"]
 
(* <|"id.0001" -> "1604", "id.0002" -> "THE TRAGEDY OF HAMLET, PRINCE OF DENMARK", <<220>>, "id.0223" -> "THE END"|> *) 

Another generic function from [AAp1] is LSAMonTakeValue (used many times above.)

Here is an example of the “data dropper” LSAMonDropDocuments:

Keys[p⟹LSAMonDropDocuments⟹QRMonTakeContext]

(* {"documentTermMatrix", "terms", "stopWords", "stemmingRules"} *)

(The “droppers” simply use the state monad function LSAMonDropFromContext, [AAp1]. For example, LSAMonDropDocuments is equivalent to LSAMonDropFromContext[“documents”].)

The utilization of SSparseMatrix objects

The LSAMon monad heavily relies on SSparseMatrix objects, [AAp6, AA5], for internal representation of data and computation results.

A SSparseMatrix object is a matrix with named rows and columns.

Here is an example.

n = 6;
rmat = ToSSparseMatrix[
   SparseArray[{{1, 2} -> 1, {4, 5} -> 1}, {n, n}], 
   "RowNames" -> RandomSample[CharacterRange["A", "Z"], n], 
   "ColumnNames" -> RandomSample[CharacterRange["a", "z"], n]];
MatrixForm[rmat]
LSAMon-The-utilization-of-SSparseMatrix-random-matrix
LSAMon-The-utilization-of-SSparseMatrix-random-matrix

In this section we look into some useful SSparseMatrix idioms applied within LSAMon.

Visualize with sorted rows and columns

In some situations it is beneficial to sort rows and columns of the (weighted) document-term matrix.

docTermMat = 
  lsaSpeeches⟹LSAMonTakeDocumentTermMatrix;
MatrixPlot[docTermMat[[Ordering[RowSums[docTermMat]],  Ordering[ColumnSums[docTermMat]]]], MaxPlotPoints -> 300, ImageSize -> Large]
LSAMon-The-utilization-of-SSparseMatrix-lsaSpeeces-docTermMat-plot
LSAMon-The-utilization-of-SSparseMatrix-lsaSpeeces-docTermMat-plot

The most popular terms in the document collection can be found through the association of the column sums of the document-term matrix.

TakeLargest[ColumnSumsAssociation[lsaSpeeches⟹LSAMonTakeDocumentTermMatrix], 10]

(* <|"state" -> 8852, "govern" -> 8147, "year" -> 6362, "nation" -> 6182,
     "congress" -> 5040, "unit" -> 5040, "countri" -> 4504, 
     "peopl" -> 4306, "american" -> 3648, "law" -> 3496|> *)
     

Similarly for the lest popular terms.

TakeSmallest[
 ColumnSumsAssociation[
  lsaSpeeches⟹LSAMonTakeDocumentTermMatrix], 10]

(* <|"036" -> 1, "027" -> 1, "_____________________" -> 1, "0111" -> 1, 
     "006" -> 1, "0000" -> 1, "0001" -> 1, "______________________" -> 1, 
     "____" -> 1, "____________________" -> 1|> *)

Showing only non-zero columns

In some cases we want to show only columns of the data or computation results matrices that have non-zero elements.

Here is an example (similar to other examples in the previous section.)

lsaHamlet⟹
  LSAMonRepresentByTerms[{"this country is rotten", 
    "where is my sword my lord", 
    "poison in the ear should be in the play"}]⟹
  LSAMonEchoFunctionValue[ MatrixForm[#1[[All, Keys[Select[ColumnSumsAssociation[#1], #1 > 0 &]]]]] &];
LSAMon-The-utilization-of-SSparseMatrix-lsaHamlet-queries-to-terms-matrix
LSAMon-The-utilization-of-SSparseMatrix-lsaHamlet-queries-to-terms-matrix

In the pipeline code above: (i) from the list of queries a representation matrix is made, (ii) that matrix is assigned to the pipeline value, (iii) in the pipeline echo value function the non-zero columns are selected with by using the keys of the non-zero elements of the association obtained with ColumnSumsAssociation.

Similarities based on representation by terms

Here is way to compute the similarity matrix of different sets of documents that are not required to be in monad’s document collection.

sMat1 =
 lsaSpeeches⟹
  LSAMonRepresentByTerms[ aStateOfUnionSpeeches[[ Range[-5, -2] ]] ]⟹
  LSAMonTakeValue

sMat2 =
 lsaSpeeches⟹
  LSAMonRepresentByTerms[ aStateOfUnionSpeeches[[ Range[-7, -3] ]] ]⟹
  LSAMonTakeValue

MatrixForm[sMat1.Transpose[sMat2]]
LSAMon-The-utilization-of-SSparseMatrix-lsaSpeeches-terms-similarities-matrix
LSAMon-The-utilization-of-SSparseMatrix-lsaSpeeches-terms-similarities-matrix

Similarities based on representation by topics

Similarly to weighted Boolean similarities matrix computation above we can compute a similarity matrix using the topics representations. Note that an additional normalization steps is required.

sMat1 =
  lsaSpeeches⟹
   LSAMonRepresentByTopics[ aStateOfUnionSpeeches[[ Range[-5, -2] ]] ]⟹
   LSAMonTakeValue;
sMat1 = WeightTermsOfSSparseMatrix[sMat1, "None", "None", "Cosine"]

sMat2 =
  lsaSpeeches⟹
   LSAMonRepresentByTopics[ aStateOfUnionSpeeches[[ Range[-7, -3] ]] ]⟹ 
   LSAMonTakeValue;
sMat2 = WeightTermsOfSSparseMatrix[sMat2, "None", "None", "Cosine"]

MatrixForm[sMat1.Transpose[sMat2]]
LSAMon-The-utilization-of-SSparseMatrix-lsaSpeeches-topics-similarities-matrix
LSAMon-The-utilization-of-SSparseMatrix-lsaSpeeches-topics-similarities-matrix

Note the differences with the weighted Boolean similarity matrix in the previous sub-section – the similarities that are less than 1 are noticeably larger.

Unit tests

The development of LSAMon was done with two types of unit tests: (i) directly specified tests, [AAp7], and (ii) tests based on randomly generated pipelines, [AA8].

The unit test package should be further extended in order to provide better coverage of the functionalities and illustrate – and postulate – pipeline behavior.

Directly specified tests

Here we run the unit tests file “MonadicLatentSemanticAnalysis-Unit-Tests.wlt”, [AAp8].

AbsoluteTiming[
 testObject = TestReport["~/MathematicaForPrediction/UnitTests/MonadicLatentSemanticAnalysis-Unit-Tests.wlt"]
]

The natural language derived test ID’s should give a fairly good idea of the functionalities covered in [AAp3].

Values[Map[#["TestID"] &, testObject["TestResults"]]]

(* {"LoadPackage", "USASpeechesData", "HamletData", "StopWords", 
    "Make-document-term-matrix-1", "Make-document-term-matrix-2",
    "Apply-term-weights-1", "Apply-term-weights-2", "Topic-extraction-1",
    "Topic-extraction-2", "Topic-extraction-3", "Topic-extraction-4",
    "Statistical-thesaurus-1", "Topics-representation-1",
    "Take-document-term-matrix-1", "Take-weighted-document-term-matrix-1",
    "Take-document-term-matrix-2", "Take-weighted-document-term-matrix-2",
    "Take-terms-1", "Take-Factors-1", "Take-Factors-2", "Take-Factors-3",
    "Take-Factors-4", "Take-StopWords-1", "Take-StemmingRules-1"} *)

Random pipelines tests

Since the monad LSAMon is a DSL it is natural to test it with a large number of randomly generated “sentences” of that DSL. For the LSAMon DSL the sentences are LSAMon pipelines. The package “MonadicLatentSemanticAnalysisRandomPipelinesUnitTests.m”, [AAp9], has functions for generation of LSAMon random pipelines and running them as verification tests. A short example follows.

Generate pipelines:

SeedRandom[234]
pipelines = MakeLSAMonRandomPipelines[100];
Length[pipelines]

(* 100 *)

Here is a sample of the generated pipelines:

LSAMon-Unit-tests-random-pipelines-sample-table
LSAMon-Unit-tests-random-pipelines-sample-table

Here we run the pipelines as unit tests:

AbsoluteTiming[
 res = TestRunLSAMonPipelines[pipelines, "Echo" -> False];
]

From the test report results we see that a dozen tests failed with messages, all of the rest passed.

rpTRObj = TestReport[res]

(The message failures, of course, have to be examined – some bugs were found in that way. Currently the actual test messages are expected.)

Future plans

Dimension reduction extensions

It would be nice to extend the Dimension reduction functionalities of LSAMon to include other algorithms like Independent Component Analysis (ICA), [Wk5]. Ideally with LSAMon we can do comparisons between SVD, NNMF, and ICA like the image de-nosing based comparison explained in [AA8].

Another direction is to utilize Neural Networks for the topic extraction and making of statistical thesauri.

Conversational agent

Since LSAMon is a DSL it can be relatively easily interfaced with a natural language interface.

Here is an example of natural language commands parsed into LSA code using the package [AAp13].

LSAMon-Future-directions-parsed-LSA-commands-table
LSAMon-Future-directions-parsed-LSA-commands-table

Implementation notes

The implementation methodology of the LSAMon monad packages [AAp3, AAp9] followed the methodology created for the ClCon monad package [AAp10, AA6]. Similarly, this document closely follows the structure and exposition of the `ClCon monad document “A monad for classification workflows”, [AA6].

A lot of the functionalities and signatures of LSAMon were designed and programed through considerations of natural language commands specifications given to a specialized conversational agent.

References

Packages

[AAp1] Anton Antonov, State monad code generator Mathematica package, (2017), MathematicaForPrediction at GitHub*.

[AAp2] Anton Antonov, Monadic tracing Mathematica package, (2017), MathematicaForPrediction at GitHub*.

[AAp3] Anton Antonov, Monadic Latent Semantic Analysis Mathematica package, (2017), MathematicaForPrediction at GitHub.

[AAp4] Anton Antonov, Implementation of document-term matrix construction and re-weighting functions in Mathematica, (2013), MathematicaForPrediction at GitHub.

[AAp5] Anton Antonov, Non-Negative Matrix Factorization algorithm implementation in Mathematica, (2013), MathematicaForPrediction at GitHub.

[AAp6] Anton Antonov, SSparseMatrix Mathematica package, (2018), MathematicaForPrediction at GitHub.

[AAp7] Anton Antonov, MathematicaForPrediction utilities, (2014), MathematicaForPrediction at GitHub.

[AAp8] Anton Antonov, Monadic Latent Semantic Analysis unit tests, (2019), MathematicaVsR at GitHub.

[AAp9] Anton Antonov, Monadic Latent Semantic Analysis random pipelines Mathematica unit tests, (2019), MathematicaForPrediction at GitHub.

[AAp10] Anton Antonov, Monadic contextual classification Mathematica package, (2017), MathematicaForPrediction at GitHub.

[AAp11] Anton Antonov, Heatmap plot Mathematica package, (2017), MathematicaForPrediction at GitHub.

[AAp12] Anton Antonov,
Independent Component Analysis Mathematica package, MathematicaForPrediction at GitHub.

[AAp13] Anton Antonov, Latent semantic analysis workflows grammar in EBNF, (2018), ConverasationalAgents at GitHub.

MathematicaForPrediction articles

[AA1] Anton Antonov, “Monad code generation and extension”, (2017), MathematicaForPrediction at GitHub.

[AA2] Anton Antonov, “Topic and thesaurus extraction from a document collection”, (2013), MathematicaForPrediction at GitHub.

[AA3] Anton Antonov, “The Great conversation in USA presidential speeches”, (2017), MathematicaForPrediction at WordPress.

[AA4] Anton Antonov, “Contingency tables creation examples”, (2016), MathematicaForPrediction at WordPress.

[AA5] Anton Antonov, “RSparseMatrix for sparse matrices with named rows and columns”, (2015), MathematicaForPrediction at WordPress.

[AA6] Anton Antonov, “A monad for classification workflows”, (2018), MathematicaForPrediction at WordPress.

[AA7] Anton Antonov, “Independent component analysis for multidimensional signals”, (2016), MathematicaForPrediction at WordPress.

[AA8] Anton Antonov, “Comparison of PCA, NNMF, and ICA over image de-noising”, (2016), MathematicaForPrediction at WordPress.

Other

[Wk1] Wikipedia entry, Monad,

[Wk2] Wikipedia entry, Latent semantic analysis,

[Wk3] Wikipedia entry, Distributional semantics,

[Wk4] Wikipedia entry, Non-negative matrix factorization,

[LE1] Lars Elden, Matrix Methods in Data Mining and Pattern Recognition, 2007, SIAM. ISBN-13: 978-0898716269.

[MB1] Michael W. Berry & Murray Browne, Understanding Search Engines: Mathematical Modeling and Text Retrieval, 2nd. ed., 2005, SIAM. ISBN-13: 978-0898715811.

[MS1] Magnus Sahlgren, “The Distributional Hypothesis”, (2008), Rivista di Linguistica. 20 (1): 33[Dash]53.

[PS1] Patrick Scheibe, Mathematica (Wolfram Language) support for IntelliJ IDEA, (2013-2018), Mathematica-IntelliJ-Plugin at GitHub.

Comparison of dimension reduction algorithms over mandala images generation

Introduction

This document discusses concrete algorithms for two different approaches of generation of mandala images, [1]: direct construction with graphics primitives, and use of machine learning algorithms.

In the experiments described in this document better results were obtained with the direct algorithms. The direct algorithms were made for the Mathematica StackExchange question "Code that generates a mandala", [3].

The main goals of this document are:

  1. to show some pretty images exploiting symmetry and multiplicity (see this album),

  2. to provide an illustrative example of comparing dimension reduction methods,

  3. to give a set-up for further discussions and investigations on mandala creation with machine learning algorithms.

Two direct construction algorithms are given: one uses "seed" segment rotations, the other superimposing of layers of different types. The following plots show the order in which different mandala parts are created with each of the algorithms.

"Direct-Mandala-creation-algorithms-steps"

In this document we use several algorithms for dimension reduction applied to collections of images following the procedure described in [4,5]. We are going to show that with Non-Negative Matrix Factorization (NNMF) we can use mandalas made with the seed segment rotation algorithm to extract layer types and superimpose them to make colored mandalas. Using the same approach with Singular Value Decomposition (SVD) or Independent Component Analysis (ICA) does not produce good layers and the superimposition produces more "watered-down", less diverse mandalas.

From a more general perspective this document compares the statistical approach of "trying to see without looking" with the "direct simulation" approach. Another perspective is the creation of "design spaces"; see [6].

The idea of using machine learning algorithms is appealing because there is no need to make the mental effort of understanding, discerning, approximating, and programming the principles of mandala creation. We can "just" use a large collection of mandala images and generate new ones using the "internal knowledge" data of machine learning algorithms. For example, a Neural network system like Deep Dream, [2], might be made to dream of mandalas.

Direct algorithms for mandala generation

In this section we present two different algorithms for generating mandalas. The first sees a mandala as being generated by rotation of a "seed" segment. The second sees a mandala as being generated by different component layers. For other approaches see [3].

The request of [3] is for generation of mandalas for coloring by hand. That is why the mandala generation algorithms are in the grayscale space. Coloring the generated mandala images is a secondary task.

By seed segment rotations

One way to come up with mandalas is to generate a segment and then by appropriate number of rotations to produce a mandala.

Here is a function and an example of random segment (seed) generation:

Clear[MakeSeedSegment]
MakeSeedSegment[radius_, angle_, n_Integer: 10, 
   connectingFunc_: Polygon, keepGridPoints_: False] :=
  Block[{t},
   t = Table[
     Line[{radius*r*{Cos[angle], Sin[angle]}, {radius*r, 0}}], {r, 0, 1, 1/n}];
   Join[If[TrueQ[keepGridPoints], t, {}], {GrayLevel[0.25], 
     connectingFunc@RandomSample[Flatten[t /. Line[{x_, y_}] :> {x, y}, 1]]}]
   ];

seed = MakeSeedSegment[10, Pi/12, 10];
Graphics[seed, Frame -> True]
"Mandala-seed-segment"

This function can make a seed segment symmetric:

Clear[MakeSymmetric]
MakeSymmetric[seed_] := {seed, 
   GeometricTransformation[seed, ReflectionTransform[{0, 1}]]};

seed = MakeSymmetric[seed];
Graphics[seed, Frame -> True]
"Mandala-seed-segment-symmetric"

Using a seed we can generate mandalas with different specification signatures:

Clear[MakeMandala]
MakeMandala[opts : OptionsPattern[]] :=      
  MakeMandala[
   MakeSymmetric[
    MakeSeedSegment[20, Pi/12, 12, 
     RandomChoice[{Line, Polygon, BezierCurve, 
       FilledCurve[BezierCurve[#]] &}], False]], Pi/6, opts];

MakeMandala[seed_, angle_?NumericQ, opts : OptionsPattern[]] :=      
  Graphics[GeometricTransformation[seed, 
    Table[RotationMatrix[a], {a, 0, 2 Pi - angle, angle}]], opts];

This code randomly selects symmetricity and seed generation parameters (number of concentric circles, angles):

SeedRandom[6567]
n = 12;
Multicolumn@
 MapThread[
  Image@If[#1,
     MakeMandala[MakeSeedSegment[10, #2, #3], #2],
     MakeMandala[
      MakeSymmetric[MakeSeedSegment[10, #2, #3, #4, False]], 2 #2]
     ] &, {RandomChoice[{False, True}, n], 
   RandomChoice[{Pi/7, Pi/8, Pi/6}, n], 
   RandomInteger[{8, 14}, n], 
   RandomChoice[{Line, Polygon, BezierCurve, 
     FilledCurve[BezierCurve[#]] &}, n]}]
"Seed-segment-rotation-mandalas-complex-settings"

Here is a more concise way to generate symmetric segment mandalas:

Multicolumn[Table[Image@MakeMandala[], {12}], 5]
"Seed-segment-rotation-mandalas-simple-settings"

Note that with this approach the programming of the mandala coloring is not that trivial — weighted blending of colorized mandalas is the easiest thing to do. (Shown below.)

By layer types

This approach was given by Simon Woods in [3].

"For this one I’ve defined three types of layer, a flower, a simple circle and a ring of small circles. You could add more for greater variety."

The coloring approach with image blending given below did not work well for this algorithm, so I modified the original code in order to produce colored mandalas.

ClearAll[LayerFlower, LayerDisk, LayerSpots, MandalaByLayers]

LayerFlower[n_, a_, r_, colorSchemeInd_Integer] := 
  Module[{b = RandomChoice[{-1/(2 n), 0}]}, {If[
     colorSchemeInd == 0, White, 
     RandomChoice[ColorData[colorSchemeInd, "ColorList"]]], 
    Cases[ParametricPlot[
      r (a + Cos[n t])/(a + 1) {Cos[t + b Sin[2 n t]], Sin[t + b Sin[2 n t]]}, {t, 0, 2 Pi}], 
     l_Line :> FilledCurve[l], -1]}];

LayerDisk[_, _, r_, colorSchemeInd_Integer] := {If[colorSchemeInd == 0, White, 
    RandomChoice[ColorData[colorSchemeInd, "ColorList"]]], 
   Disk[{0, 0}, r]};

LayerSpots[n_, a_, r_, colorSchemeInd_Integer] := {If[colorSchemeInd == 0, White, 
    RandomChoice[ColorData[colorSchemeInd, "ColorList"]]], 
   Translate[Disk[{0, 0}, r a/(4 n)], r CirclePoints[n]]};

MandalaByLayers[n_, m_, coloring : (False | True) : False, opts : OptionsPattern[]] := 
  Graphics[{EdgeForm[Black], White, 
    Table[RandomChoice[{3, 2, 1} -> {LayerFlower, LayerDisk, LayerSpots}][n, RandomReal[{3, 5}], i, 
       If[coloring, RandomInteger[{1, 17}], 0]]~Rotate~(Pi i/n), {i, m, 1, -1}]}, opts];

Here are generated black-and-white mandalas.

SeedRandom[6567]
ImageCollage[Table[Image@MandalaByLayers[16, 20], {12}], Background -> White, ImagePadding -> 3, ImageSize -> 1200]
"Layer-types-superimposing-BW"

Here are some colored mandalas. (Which make me think more of Viking and Native American art than mandalas.)

ImageCollage[Table[Image@MandalaByLayers[16, 20, True], {12}], Background -> White, ImagePadding -> 3, ImageSize -> 1200]
"Layer-types-superimposing-colored"

Training data

Images by direct generation

iSize = 400;

SeedRandom[6567]
AbsoluteTiming[
 mandalaImages = 
   Table[Image[
     MakeMandala[
      MakeSymmetric@
       MakeSeedSegment[10, Pi/12, 12, RandomChoice[{Polygon, FilledCurve[BezierCurve[#]] &}]], Pi/6], 
     ImageSize -> {iSize, iSize}, ColorSpace -> "Grayscale"], {300}];
 ]

(* {39.31, Null} *)

ImageCollage[ColorNegate /@ RandomSample[mandalaImages, 12], Background -> White, ImagePadding -> 3, ImageSize -> 400]
"mandalaImages-sample"

External image data

See the section "Using World Wide Web images".

Direct blending

The most interesting results are obtained with the image blending procedure coded below over mandala images generated with the seed segment rotation algorithm.

SeedRandom[3488]
directBlendingImages = Table[
   RemoveBackground@
    ImageAdjust[
     Blend[Colorize[#, 
         ColorFunction -> 
          RandomChoice[{"IslandColors", "FruitPunchColors", 
            "AvocadoColors", "Rainbow"}]] & /@ 
       RandomChoice[mandalaImages, 4], RandomReal[1, 4]]], {36}];

ImageCollage[directBlendingImages, Background -> White, ImagePadding -> 3, ImageSize -> 1200]

"directBlendingImages-3488-36"

Dimension reduction algorithms application

In this section we are going to apply the dimension reduction algorithms Singular Value Decomposition (SVD), Independent Component Analysis (ICA), and Non-Negative Matrix Factorization (NNMF) to a linear vector space representation (a matrix) of an image dataset. In the next section we use the bases generated by those algorithms to make mandala images.
We are going to use the packages [7,8] for ICA and NNMF respectively.


Import["https://raw.githubusercontent.com/antononcube/MathematicaForPrediction/master/IndependentComponentAnalysis.m"]
Import["https://raw.githubusercontent.com/antononcube/MathematicaForPrediction/master/NonNegativeMatrixFactorization.m"]

Linear vector space representation

The linear vector space representation of the images is simple — each image is flattened to a vector (row-wise), and the image vectors are put into a matrix.

mandalaMat = Flatten@*ImageData@*ColorNegate /@ mandalaImages;
Dimensions[mandalaMat]

(* {300, 160000} *)

Re-factoring and basis images

The following code re-factors the images matrix with SVD, ICA, and NNMF and extracts the basis images.

AbsoluteTiming[
 svdRes = SingularValueDecomposition[mandalaMat, 20];
]
(* {5.1123, Null} *)

svdBasisImages = Map[ImageAdjust@Image@Partition[#, iSize] &, Transpose@svdRes[[3]]];

AbsoluteTiming[
 icaRes = 
   IndependentComponentAnalysis[Transpose[mandalaMat], 20, 
    PrecisionGoal -> 4, "MaxSteps" -> 100];
]
(* {23.41, Null} *)

icaBasisImages = Map[ImageAdjust@Image@Partition[#, iSize] &, Transpose[icaRes[[1]]]];

SeedRandom[452992]
AbsoluteTiming[
 nnmfRes = 
   GDCLS[mandalaMat, 20, PrecisionGoal -> 4, 
    "MaxSteps" -> 20, "RegularizationParameter" -> 0.1];
 ]
(* {233.209, Null} *)

nnmfBasisImages = Map[ImageAdjust@Image@Partition[#, iSize] &, nnmfRes[[2]]];

Bases

Let us visualize the bases derived with the matrix factorization methods.

Grid[{{"SVD", "ICA", "NNMF"},
      Map[ImageCollage[#, Automatic, {400, 500}, 
        Background -> LightBlue, ImagePadding -> 5, ImageSize -> 350] &, 
      {svdBasisImages, icaBasisImages, nnmfBasisImages}]
     }, Dividers -> All]
"Mandala-SVD-ICA-NNMF-bases-20"

"Mandala-SVD-ICA-NNMF-bases-20"

Here are some observations for the bases.

  1. The SVD basis has an average mandala image as its first vector and the other vectors are "differences" to be added to that first vector.

  2. The SVD and ICA bases are structured similarly. That is because ICA and SVD are both based on orthogonality — ICA factorization uses an orthogonality criteria based on Gaussian noise properties (which is more relaxed than SVD’s standard orthogonality criteria.)

  3. As expected, the NNMF basis images have black background because of the enforced non-negativity. (Black corresponds to 0, white to 1.)

  4. Compared to the SVD and ICA bases the images of the NNMF basis are structured in a radial manner. This can be demonstrated using image binarization.

Grid[{{"SVD", "ICA", "NNMF"}, Map[ImageCollage[Binarize[#, 0.5] & /@ #, Automatic, {400, 500}, Background -> LightBlue, ImagePadding -> 5, ImageSize -> 350] &, {svdBasisImages, icaBasisImages, nnmfBasisImages}] }, Dividers -> All]
"Mandala-SVD-ICA-NNMF-bases-binarized-0.5-20"

We can see that binarizing of the NNMF basis images shows them as mandala layers. In other words, using NNMF we can convert the mandalas of the seed segment rotation algorithm into mandalas generated by an algorithm that superimposes layers of different types.

Blending with image bases samples

In this section we just show different blending images using the SVD, ICA, and NNMF bases.

Blending function definition

ClearAll[MandalaImageBlending]
Options[MandalaImageBlending] = {"BaseImage" -> {}, "BaseImageWeight" -> Automatic, "PostBlendingFunction" -> (RemoveBackground@*ImageAdjust)};
MandalaImageBlending[basisImages_, nSample_Integer: 4, n_Integer: 12, opts : OptionsPattern[]] :=      
  Block[{baseImage, baseImageWeight, postBlendingFunc, sImgs, sImgWeights},
   baseImage = OptionValue["BaseImage"];
   baseImageWeight = OptionValue["BaseImageWeight"];
   postBlendingFunc = OptionValue["PostBlendingFunction"];
   Table[(
     sImgs = 
      Flatten@Join[{baseImage}, RandomSample[basisImages, nSample]];
     If[NumericQ[baseImageWeight] && ImageQ[baseImage],
      sImgWeights = 
       Join[{baseImageWeight}, RandomReal[1, Length[sImgs] - 1]],
      sImgWeights = RandomReal[1, Length[sImgs]]
      ];
     postBlendingFunc@
      Blend[Colorize[#, 
          DeleteCases[{opts}, ("BaseImage" -> _) | ("BaseImageWeight" -> _) | ("PostBlendingFunction" -> _)],               
          ColorFunction -> 
           RandomChoice[{"IslandColors", "FruitPunchColors", 
             "AvocadoColors", "Rainbow"}]] & /@ sImgs, 
       sImgWeights]), {n}]
   ];

SVD image basis blending

SeedRandom[17643]
svdBlendedImages = MandalaImageBlending[Rest@svdBasisImages, 4, 24];
ImageCollage[svdBlendedImages, Background -> White, ImagePadding -> 3, ImageSize -> 1200]

"svdBlendedImages-17643-24"

SeedRandom[17643]
svdBlendedImages = MandalaImageBlending[Rest@svdBasisImages, 4, 24, "BaseImage" -> First[svdBasisImages], "BaseImageWeight" -> 0.5];
ImageCollage[svdBlendedImages, Background -> White, ImagePadding -> 3, ImageSize -> 1200]

"svdBlendedImages-baseImage-17643-24"

ICA image basis blending

SeedRandom[17643]
icaBlendedImages = MandalaImageBlending[Rest[icaBasisImages], 4, 36, "BaseImage" -> First[icaBasisImages], "BaseImageWeight" -> Automatic];
ImageCollage[icaBlendedImages, Background -> White, ImagePadding -> 3, ImageSize -> 1200]

"icaBlendedImages-17643-36"

NNMF image basis blending

SeedRandom[17643]
nnmfBlendedImages = MandalaImageBlending[nnmfBasisImages, 4, 36];
ImageCollage[nnmfBlendedImages, Background -> White, ImagePadding -> 3, ImageSize -> 1200]

"nnmfBlendedImages-17643-36"

Using World Wide Web images

A natural question to ask is:

What would be the outcomes of the above procedures to mandala images found in the World Wide Web (WWW) ?

Those WWW images are most likely man made or curated.

The short answer is that the results are not that good. Better results might be obtained using a larger set of WWW images (than just 100 in the experiment results shown below.)

Here is a sample from the WWW mandala images:

"wwwMandalaImages-sample-6

Here are the results obtained with NNMF basis:

"www-nnmfBlendedImages-12"

Future plans

My other motivation for writing this document is to set up a basis for further investigations and discussions on the following topics.

  1. Having a large image database of "real world", human made mandalas.

  2. Utilization of Neural Network algorithms to mandala creation.

  3. Utilization of Cellular Automata to mandala generation.

  4. Investigate mandala morphing and animations.

  5. Making a domain specific language of specifications for mandala creation and modification.

The idea of using machine learning algorithms for mandala image generation was further supported by an image classifier that recognizes fairly well (suitably normalized) mandala images obtained in different ways:

"Mandalas-classifer-measurements-matrix"

References

[1] Wikipedia entry: Mandala, https://en.wikipedia.org/wiki/Mandala .

[2] Wikipedia entry: DeepDream, https://en.wikipedia.org/wiki/DeepDream .

[3] "Code that generates a mandala", Mathematica StackExchange, http://mathematica.stackexchange.com/q/136974 .

[4] Anton Antonov, "Comparison of PCA and NNMF over image de-noising", (2016), MathematicaForPrediction at WordPress blog. URL: https://mathematicaforprediction.wordpress.com/2016/05/07/comparison-of-pca-and-nnmf-over-image-de-noising/ .

[5] Anton Antonov, "Handwritten digits recognition by matrix factorization", (2016), MathematicaForPrediction at WordPress blog. URL: https://mathematicaforprediction.wordpress.com/2016/11/12/handwritten-digits-recognition-by-matrix-factorization/ .

[6] Chris Carlson, "Social Exploration of Design Spaces: A Proposal", (2016), Wolfram Technology Conference 2016. URL: http://wac .36f4.edgecastcdn.net/0036F4/pub/www.wolfram.com/technology-conference/2016/SocialExplorationOfDesignSpaces.nb , YouTube: https://www.youtube.com/watch?v=YK2523nfcms .

[7] Anton Antonov, Independent Component Analysis Mathematica package, (2016), source code at MathematicaForPrediction at GitHub, package IndependentComponentAnalysis.m .

[8] Anton Antonov, Implementation of the Non-Negative Matrix Factorization algorithm in Mathematica, (2013), source code at MathematicaForPrediction at GitHub, package NonNegativeMatrixFactorization.m.

Handwritten digits recognition by matrix factorization

Introduction

This MathematicaVsR at GitHub project is for comparing Mathematica and R for the tasks of classifier creation, execution, and evaluation using the MNIST database of images of handwritten digits.

Here are the bases built with two different classifiers:

  • Singular Value Decomposition (SVD)

SVD-basis-for-5

  • Non-Negative Matrix Factorization (NNMF)

NNMF-basis-for-5

Here are the confusion matrices of the two classifiers:

  • SVD

SVD-confusion-matrix

  • NNMF

NNMF-confusion-matrix

The blog post "Classification of handwritten digits" (published 2013) has a related more elaborated discussion over a much smaller database of handwritten digits.

Concrete steps

The concrete steps taken in scripts and documents of this project follow.

  1. Ingest the binary data files into arrays that can be visualized as digit images.
  • We have two sets: 60,000 training images and 10,000 testing images.
  1. Make a linear vector space representation of the images by simple unfolding.

  2. For each digit find the corresponding representation matrix and factorize it.

  3. Store the matrix factorization results in a suitable data structure. (These results comprise the classifier training.)

  • One of the matrix factors is seen as a new basis.
  1. For a given test image (and its linear vector space representation) find the basis that approximates it best. The corresponding digit is the classifier prediction for the given test image.

  2. Evaluate the classifier(s) over all test images and compute accuracy, F-Scores, and other measures.

Scripts

There are scripts going through the steps listed above:

Documents

The following documents give expositions that are suitable for reading and following of steps and corresponding results.

Observations

Ingestion

I figured out first in R how to ingest the data in the binary files of the MNIST database. There were at least several online resources (blog posts, GitHub repositories) that discuss the MNIST binary files ingestion.

After that making the corresponding code in Mathematica was easy.

Classification results

Same in Mathematica and R for for SVD and NNMF. (As expected.)

NNMF

NNMF classifiers use the MathematicaForPrediction at GitHub implementations: NonNegativeMatrixFactorization.m and NonNegativeMatrixFactorization.R.

Parallel computations

Both Mathematica and R have relatively simple set-up of parallel computations.

Graphics

It was not very straightforward to come up in R with visualizations for MNIST images. The Mathematica visualization is much more flexible when it comes to plot labeling.

Going further

Comparison with other classifiers

Using Mathematica’s built-in classifiers it was easy to compare the SVD and NNMF classifiers with neural network ones and others. (The SVD and NNMF are much faster to built and they bring comparable precision.)

It would be nice to repeat that in R using one or several of the neural network classifiers provided by Google, Microsoft, H2O, Baidu, etc.

Classifier ensembles

Another possible extension is to use classifier ensembles and Receiver Operation Characteristic (ROC) to create better classifiers. (Both in Mathematica and R.)

Importance of variables

Using classifier agnostic importance of variables procedure we can figure out :

  • which NNMF basis vectors (images) are most important for the classification precision,

  • which image rows or columns are most important for each digit, or similarly

  • which image squares of a, say, 4×4 image grid are most important.