8000 GitHub - duytinvo/acl2016: Don't Count, Predict! An Automatic Approach to Learning Sentiment Lexicons for Short Text · GitHub
[go: up one dir, main page]

Skip to content
duytinvoPublic

About

Don't Count, Predict! An Automatic Approach to Learning Sentiment Lexicons for Short Text

Resources

Stars

13 stars

Watchers

3 watching

Forks

Latest commit

 

History

9 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 

Repository files navigation

acl2016

This code is used for the paper "Don't Count, Predict! An Automatic Approach to Learning Sentiment Lexicons for Short Text" (ACL2016-short paper)

This repository consists of three folders:

  • learn_lexicon: scripts to learn lexicons
  • test_lexicon: scripts to test ours learned lexicons
  • lexicons: sentiment lexicons in English and Arabic, which are learned by our models.

To use our lexicons:

  • English and Arabic lexicons in "lexicons" folder.
  • Data format: word + '\t' + sentiment score + '\n'

To compare our lexicons with available lexicons:

  • Change current directory to folder "acl2016/test_lexicon/scripts"
  • Change all *.sh file to 755 (e.g. chmod 755 *.sh)
  • Change all files in folder liblinear to 755 (e.g. chmod 755 ./liblinear/*)
  • For testing English lexicons:
    • Extract features: ./run_features.sh
    • Classification: ./run_liblinear.sh
  • For testing Arabic lexicons:
    • Extract features: ./run_features_ar.sh
    • Classification: ./run_liblinear_ar.sh

To run the our models to learn lexicons:

TO LEARN MODEL BY YOUR OWN DATA:

  • Process your data:
    • Example: python process.py ../data/alexgo/raw/metatweets ../data/alexgo/processed/info.tw ../data/alexgo/processed/process.tw
    • Raw data format: labels (1 or 0) + ' ' + tweet +'\n'
  • Modify the script by changing the inputs to ../data/alexgo/processed/info.tw and ../data/alexgo/processed/process.tw
  • Run model

Reference:

@InProceedings{vo-zhang:2016:acl,

author = {Vo, Duy Tin and Zhang, Yue},

title = {Don't Count, Predict! An Automatic Approach to Learning Sentiment Lexicons for Short Text},

booktitle = {Proceedings of ACL},

month = {August},

year = {2016},

address = {Berlin, Germany}}

About

Don't Count, Predict! An Automatic Approach to Learning Sentiment Lexicons for Short Text

Resources

Stars

13 stars

Watchers

3 watching

Forks

Releases

Packages

Contributors

Languages

0