/CI_Text_Classification

Classifying an extremely large Persian news dataset by using neural network algorithms written in python.

Primary LanguageJupyter Notebook

CI_Text_Recognition

Classifying an extremely large Persian news dataset by using neural network algorithms written in python.

There are two implementations here:

1- Tensorflow Keras implementation (GRU Neural Network)

2- SKLearn implementation (SGDClassifier)

The preprocessing algorithm uses opensource project HAZM,

https://github.com/sobhe/hazm

and the persian stopwords are collected from the following library

https://github.com/kharazi/persian-stopwords

Dataset is not available here due to copyright law.

(SBU Computational Intelligence Course - Final Project)