GitHub - rahuljha/content-models: Implementation of probabilistic LDA style content models

rahuljha / content-models Public

Notifications You must be signed in to change notification settings
Fork 0
Star 3

Implementation of probabilistic LDA style content models

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 8 Commits
Data.py		Data.py
Readme.txt		Readme.txt
SurveyorData.py		SurveyorData.py
TopicSum.py		TopicSum.py
TopicSumNumpy.py		TopicSumNumpy.py
TopicSumWeave.py		TopicSumWeave.py
compute_hits.py		compute_hits.py
lexrank.py		lexrank.py

Repository files navigation

Relevant code for the following publication:

Rahul Jha, Catherine Finegan-Dollak, Ben King, Reed Coke and Dragomir Radev. Content Models for Survey Generation: A Factoid-Based Evaluation. Annual Meeting of the Association for Computational Linguistics, 2015.

-----

Python implementation for TopicSum and HitSum.

TopicSum implementation is based on:

http://www.mblondel.org/journal/2010/08/21/latent-dirichlet-allocation-in-python/
https://github.com/rebeccamason/codesample/tree/master/TopicSum

HitSum code will need to be adapted to suit your needs, I have added comments wherever possible.

-----

For any questions regarding this code, please email rahuljha@umich.edu.