Nepali Text Classification
- 1 minSince there is very less research in Nepali language, I wanted to initiate some research into Nepali language starting with text classification. My goal is to work with Hierarchical Attention Network to classify the news documents written in Nepali language into various categories like Sports, Business, Politics and 13 other categories. I have learned about text preprocess and corpus creation for Nepali language.
What’s inside :-
- Corpus creation
- Dataset creation
- Text Preprocessing
- Word embeddings (word2vec)
- Character-RNN/LSTM/GRU for name generation
- Traditional machine learning algorithms:
- Logistic Regression
- kNN
- SVM
- Naive Bayes
- Ensemble Learning (Gradient Boosting, Bagging)
- Perceptron
- MLP
- Deep Learning
- RNN
- LSTM
- GRU
- Data Visualization
Images:
Here are some of the interesting results after running various machine learning algorithms.
