iDoctor: Personalized and professionalized medical recommendations based on hybrid matrix factorization
详细信息 查看全文
- 作者：Yin Zhanga ; yin.zhang.cn@ieee.org" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Min Chenb ; minchen2012@hust.edu.cn" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Dijiang Huangc ; Dijiang.Huang@asu.edu" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Di Wud ; wudi27@sysu.edu.cn" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Yong Lie ; liyong07@tsinghua.edu.cn" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae
- 关键词：Healthcare ; Recommendation ; Matrix factorization ; Sentiment analysis ; Topic model
- 刊名：Future Generation Computer Systems
- 出版年：2017
- 出版时间：January 2017
- 年：2017
- 卷：66
- 期：Complete
- 页码：30-35
- 全文大小：1390 K
文摘
We propose a topic model based approach to discover user preference distribution and doctor feature distribution.
We propose an emotion-aware approach to identify emotional offset in user reviews via sentiment analysis.
We incorporate topic model and emotional offset into the matrix factorization model.
The experimental results show that the proposed model provides a high-performance healthcare recommendation.