Identifying interesting Twitter contents using topical analysis
作者：
Highlights：
• We discover interesting tweets for a wide audience based on topic identification.
• We model Trend Sensitive-LDA that reflects the current and popular trends.
• We weight topics by exploiting their representative words.
• We weight topics by analyzing spatial and temporal variation of their probabilities.
• The most interesting tweets contain latent topics that are assigned a high weight.
摘要
•We discover interesting tweets for a wide audience based on topic identification.•We model Trend Sensitive-LDA that reflects the current and popular trends.•We weight topics by exploiting their representative words.•We weight topics by analyzing spatial and temporal variation of their probabilities.•The most interesting tweets contain latent topics that are assigned a high weight.
论文关键词：Twitter,Interesting content,Topic model,LDA,Social media
论文评审过程：Available online 16 January 2014.
论文官网地址：https://doi.org/10.1016/j.eswa.2013.12.051
