Skip to content

Latest commit

 

History

History
4 lines (3 loc) · 380 Bytes

File metadata and controls

4 lines (3 loc) · 380 Bytes

For obfuscated text, most of the NLP domain knowledge such as Word2Vec, POS-tagging, etc., are not useful since the text is impossible to read nor parse.
As a matter of fact, neural networks come in handy to create an end-to-end module that learns the embedding, feature extraction and classification.

Please refer to README.pdf for details and experiment results