This python program classifies text instances based on the lower and upper approximation of x to all the lables. This program developed based on Fuzzy Rough Nearest Neighbor (FRNN) algorithm described in the article: Jensen, R., & Cornelis, C. (2011). Fuzzy-rough nearest neighbor classification and prediction. Theoretical Computer Science, 412(42), 5871-5884 This program runnable on a command: (python)* FRNN.py datasetname.csv *optional The classification is based on 5 cross_validation classification results evaluated using the following metrics: 1.acccuracy 2.root mean squared error (RMSE)
Note: the program is not perfectly coding the alogritms but it is a replicate of the implementation on the paper. This program is subject for further improvement.
If you use the program, please cite us.