Skip to content

dsrodbae/BinRec

Repository files navigation

BinRec

A novel User-Based Collabarotive Filtering approach, implemented in C++ This software has been used for the experimentation during the research work. The parameters are the following:

  • Number of rows and columns of input dataset.
  • The training dataset, in CSV format (an example is included: ml-1m_3_90_10_Opt4.rar).
  • The biclusters dataset, in CSV format (an example is included: ml-100k_Bin_3_90_10_Opt4_biclusters.rar).
  • The Test dataset, in CSV format (an example is included:ml-100k_Bin_3_90_10_Opt4_TestPoints.csv).
  • The binarize threshold value (Sv).

About

A novel User-Based Collabarotive Filtering approach

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors