Multi-interest semantic changes over time in short-text microblogs
作者：
Highlights：
• User-representative profiling framework based on time-variational disseminated content.
• Timeseries-based user interest modelling as categorical word embedding distributions.
• Extraction of temporal semantic patterns in vectors for interests representation.
• Semantic relevance validation on a demographically relevant control set.
• Collective intelligence framework for short-text microblog third-party content providers.
摘要
•User-representative profiling framework based on time-variational disseminated content.•Timeseries-based user interest modelling as categorical word embedding distributions.•Extraction of temporal semantic patterns in vectors for interests representation.•Semantic relevance validation on a demographically relevant control set.•Collective intelligence framework for short-text microblog third-party content providers.
论文关键词：User profiling,Text mining,Neural Networks,Information retrieval,Short-text microblogs
论文评审过程：Received 28 November 2020, Revised 17 June 2021, Accepted 19 June 2021, Available online 7 July 2021, Version of Record 15 July 2021.
论文官网地址：https://doi.org/10.1016/j.knosys.2021.107249
