Mining social influence in science and vice-versa: A topic correlation approach
作者：
Highlights：
• A method for extracting topics from heterogeneous data from different sources and domains is proposed.
• Topic Similarity is addressed for comparison of topics from different backgrounds.
• Social influence in science is detected with topics from both domains.
• The proposed methods and metrics are independent from text idiom.
• Evaluation is made in the scenario of a disease outbreak, with results suggesting a cross-domain relationship.
摘要
•A method for extracting topics from heterogeneous data from different sources and domains is proposed.•Topic Similarity is addressed for comparison of topics from different backgrounds.•Social influence in science is detected with topics from both domains.•The proposed methods and metrics are independent from text idiom.•Evaluation is made in the scenario of a disease outbreak, with results suggesting a cross-domain relationship.
论文关键词：Topic modeling,Social networks,Science networks,Topic labeling,Influence mining,Topic similarity
论文评审过程：Received 11 January 2019, Revised 9 September 2019, Accepted 2 October 2019, Available online 4 December 2019, Version of Record 24 February 2020.
论文官网地址：https://doi.org/10.1016/j.ijinfomgt.2019.10.002
