Skip to content

A siamese adaptation of the Long Short-Term Memory (LSTM) network for labeled data comprised of pairs of variable-length sequences.

Notifications You must be signed in to change notification settings

amanraj209/siamese-lstm-sentence-similarity

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Siamese LSTM For Sentence Similarity

A siamese recurrent architecture to learn semantic sentence similarity is implemented using Manhattan distance as a distance metric between pairwise representation of sentences.

The model is trained and tested on Quora Question Pair Dataset with test accuracy of 82%.

Original paper: Siamese Recurrent Architectures for Learning Sentence Similarity

Quora Question Pair Dataset can be downloaded from Kaggle.

Framwork Used

  • Keras

About

A siamese adaptation of the Long Short-Term Memory (LSTM) network for labeled data comprised of pairs of variable-length sequences.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages