Topic Classification

Methodology

1.Preprocessing

2.Training

3.Testing

4.Accuracy

Preprocessing

run split.py file . It will Create a folder name Lewisplit with subfolder 1.Not used 2.Train 3.Test In these folder only body of the .sgm file will be there.

After that run stem.py file . It will Create a folder name stem with subfolder 1.Not used 2.Train 3.Test In these folder only body of the .sgm file will be there with stemming.

These dataset will be used for further use.

Training

run table.py file. It will create table.csv which will be furthur used for testing purposes.

Testing

run testing.py file . It will create testing.csv file which will have the predicted topic of the testing file.

Accuracy

run accuracy.py file. It will tell recall precission and harmonic mean of it .

link of the repository is https://github.com/mayankmahavar111/reuters