rmit-ir / AnswerPassageQuality

Code/data for SIGIR '18 paper "Ranking Documents by Answer-Passage Quality"

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

AnswerPassageQuality

This repo contains code/scripts needed to reproduce the results in the SIGIR '18 paper "Ranking Documents by Answer-Passage Quality".

If you use the data or code from this repo, please cite the following paper:

@inproceedings{yulianti_ranking_2018,
  author = {Yulianti, Evi and Chen, Ruey-Cheng and Scholer, Falk and
            Croft, W. Bruce and Sanderson, Mark},
  title = {Ranking Documents by Answer-Passage Quality},
  booktitle = {Proceedings of {SIGIR} '18},
  year = {2018},
  pages = {335--344},
  publisher = {ACM}
} 

Get Started

Install the dependencies:

pip install -r requirements.txt

Install the following packages from github repos:

Note that, the DrQA package needs to be installed via pip3 with all pretrained models downloaded:

pip3 install .
./download.sh

Ranking Experiments

Contributors

  • Ruey-Cheng Chen
  • Evi Yulianti

About

Code/data for SIGIR '18 paper "Ranking Documents by Answer-Passage Quality"

License:MIT License


Languages

Language:Roff 94.5%Language:Python 4.9%Language:Makefile 0.6%