Collaborative Expert Recommendation for Community-Based Question Answering
Abstract
With the development of Internet, users can share knowledge by asking and answering questions on community question answering (CQA) websites. How to find related experts to contribute their answers is hence worthy of studying. In this paper, we propose a recommendation algorithm called collaborative expert recommendation (CER) for this purpose. We take full advantage of the heterogeneous information including question tags, content, answer’s votes, which are considered important for identifying experts. Moreover, we combine such information by a causal assumption of questions and answers, and inner connection exploitation among different types of information such as (questioner, question), (answer, question) and (answerer, question, answer) correlations, which are more explicable and reasonable comparing with the existing methods. Experiments carried out on six real-world datasets prove that CER has a better performance.
Cite
Text
Xu et al. "Collaborative Expert Recommendation for Community-Based Question Answering." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2016. doi:10.1007/978-3-319-46128-1_24Markdown
[Xu et al. "Collaborative Expert Recommendation for Community-Based Question Answering." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2016.](https://mlanthology.org/ecmlpkdd/2016/xu2016ecmlpkdd-collaborative/) doi:10.1007/978-3-319-46128-1_24BibTeX
@inproceedings{xu2016ecmlpkdd-collaborative,
title = {{Collaborative Expert Recommendation for Community-Based Question Answering}},
author = {Xu, Congfu and Wang, Xin and Guo, Yunhui},
booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
year = {2016},
pages = {378-393},
doi = {10.1007/978-3-319-46128-1_24},
url = {https://mlanthology.org/ecmlpkdd/2016/xu2016ecmlpkdd-collaborative/}
}