Text Classification Models

This repository contains code for text multi-classification tasks. The following four models are implemented:

• Naive Bayes classifier using Bag of words feature
• Linear SVM
• Neural network
• Long short-term memory network (LSTM)

Run command

Naive Bayes and SVM

Before training naive bayes and SVM models, we need to generate new dataset files:

cd tools

python data_process.py

Train the Naive bayes model:

python nb_svm_train.py nb

Train the SVM model:

python nb_svm_train.py svm

Test a Naive bayes or SVM model:

python nb_svm_inference.py svm

SVM with word2Vec

Pay attention to the data path. Follow the steps in the 'svm_with_word2vec.ipynb'.

Neural nets

Train a simple neural network:

python nn_train.py

Test a simple neural network:

python nn_inference.py

LSTM and Explainability of LSTM

Pay attention to the data path. Follow the steps in the 'LSTM.ipynb'.