Skip to content
No description, website, or topics provided.
Python Jupyter Notebook
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

Recognizing Supportive and Unsupportive Responses


Supportr is the code release for labeling replies on the relative supportiveness, based on the paper It’s going to be okay: Measuring Access to Support in Online Communities by Zijian Wang and David Jurgens (in proceedings of EMNLP 2018).

See the project website for full details, including contact information.


The code is not yet setup as a python package but the code can be run using the script in the main directory.


The file lists the model dependencies. Aside from these python packages, the code requires the Google news word2vec vectors in the resources/ directory. The model described in the paper uses LIWC categories as features. LIWC cannot be redistributed so these categories are not included in this release. However, the Empath library is known to closely approximate the categories and the code is setup to Just Work™ if you don't have LIWC purchase. If you do have LIWC, it should be put in resources/lexicons/ and named en_liwc.txt. Additional tests are needed to report performance when LIWC is not used.


Included in this release is the aggregrated and labeled training data for the paper, which can be used to reproduce the results of the paper or improve the classifier.

Release History

  • 0.1 Initial code release

The next step is to package the code up as a module to allow easier classification. Pull requests welcome.

You can’t perform that action at this time.