Wolfram Language Paclet Repository
Community-contributed installable additions to the Wolfram Language
A software monad for Sparse Matrix Recommender (SMR) workflows
Contributed by: Anton Antonov
A software monad for constructing and using Item-Item Recommender (IIR) systems based on Sparse Matrix Algebra.
To install this paclet in your Wolfram Language environment,
evaluate this code:
PacletInstall["AntonAntonov/MonadicSparseMatrixRecommender"]
To load the code after installation, evaluate this code:
Needs["AntonAntonov`MonadicSparseMatrixRecommender`"]
Here is a dataset:
| In[1]:= | ![]() |
| Out[4]= | ![]() |
Here is a summary of the data:
| In[5]:= |
| Out[5]= | ![]() |
Here is a monadic Sparse Matrix Recommender Monad (SMRMon) pipeline for creating a recommender object and computing a recommendation by profile:
| In[6]:= | ![]() |
| Out[6]= | ![]() |
Here are the tag-type sub-matrices:
| In[7]:= |
| Out[7]= | ![]() |
Here is a recommendation by profile:
| In[8]:= | ![]() |
| Out[8]= | ![]() |
The Sparse Matrix Recommender (SMR) objects can be used to find outliers. Here is an example of finding the outlier identifiers of the Titanic data SMR object created above:
| In[9]:= | ![]() |
| Out[10]= |
Here is a summary of the found outliers:
| In[11]:= |
| Out[11]= | ![]() |
Note that both top and bottom outliers above.