Skip to content

Semantic Textual Similarity (STS) measures the degree of equivalence in the underlying semantics of paired snippets of text.

License

Notifications You must be signed in to change notification settings

zhuyuying/Semantic-Texual-Similarity-Toolkits

 
 

Repository files navigation

句子相似度

Semantic Textual Similarity Toolkits

Installation

  1. download the repo

  2. run the corenlp server standford CoreNLP 3.6.0 download

    • Usage
    $ cd stanford-corenlp-full-2015-12-09/
    $ java -mx4g -cp "*" edu.stanford.nlp.pipeline.StanfordCoreNLPServer')
    
  3. python

    • requirements.txt
    • Usage
    $ pip install -r requirements.txt
    
    • Download the NLTK stopword corpus:
    $ python -m nltk.downloader stopwords
    
  4. Run

    $python example.py
    

TODO

  • chrome driver, to caputure mt features
  • theano, lasagne, deep learning scores
    • stst/main/make_sts_nn.py
    • stst/main/make_sts_iclr.py

Embedding Feautures

  • Pre-trained Embedding

Machine Translation Features

About

Semantic Textual Similarity (STS) measures the degree of equivalence in the underlying semantics of paired snippets of text.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 97.8%
  • Java 1.1%
  • Other 1.1%