Quality Evaluation for Answers in Chinese QA Community Based on Random Forest -- Examples from Zhihu

DOI

10.2991/hsmet-17.2017.218How to use a DOI?

Keywords

QA Community, Quality Evaluation, Random Forest

Abstract

As user-generated content develops continuously, online QA communities have become a major way to access high quality knowledge and information. Common concern of these communities is therefore to determine high quality answers accurately. This paper focuses on the largest Chinese QA community website Zhihu, and studies part of answers of the twenty popular topics we choose on it. We implement feature extraction to these answers in basic information, diversity and correlation with questions, and then apply random forest classifier to model and analyze the data. The classification model can predict the quality of answers with reasonable accuracy of more than 86%. Robustness test shows that the model has high stability.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Hongwei Wang
AU  - Yuqiang Ji
AU  - Wei Wang
PY  - 2017/02
DA  - 2017/02
TI  - Quality Evaluation for Answers in Chinese QA Community Based on Random Forest -- Examples from Zhihu
BT  - Proceedings of the 2017 International Conference on Humanities Science, Management and Education Technology (HSMET 2017)
PB  - Atlantis Press
SP  - 1184
EP  - 1190
SN  - 2352-5398
UR  - https://doi.org/10.2991/hsmet-17.2017.218
DO  - 10.2991/hsmet-17.2017.218
ID  - Wang2017/02
ER  -