Mining association rules from COVID-19 related twitter data to discover word patterns, topics and inferences

作者：

Highlights：

• We review recent advances on topic and opinion extraction methods from social media data.

• We perform and assess topic extraction related with the COVID-19 use case.

• We improve Latent Dirichlet Allocation by incorporating Association Rule Mining to deal with common discrepancies.

• Our proposed methodology reduces the number of topics, whilst retaining the core themes of user attitudes.

摘要

•We review recent advances on topic and opinion extraction methods from social media data.•We perform and assess topic extraction related with the COVID-19 use case.•We improve Latent Dirichlet Allocation by incorporating Association Rule Mining to deal with common discrepancies.•Our proposed methodology reduces the number of topics, whilst retaining the core themes of user attitudes.

论文关键词：Social media,Topic extraction,Association rule mining,Data mining,COVID-19

论文评审过程：Received 5 March 2021, Revised 21 September 2021, Accepted 20 April 2022, Available online 25 April 2022, Version of Record 7 May 2022.

论文官网地址：https://doi.org/10.1016/j.is.2022.102054