Early author profiling on Twitter using profile features with multi-resolution

作者：

Highlights：

• The Early Author Profiling task is feasible in social media documents.

• Very small amounts of textual evidence can be modeled from early stages.

• The proposed meta-words capture discriminative information of target profiles.

• The proposed multi-resolution representation boosted the early recognition.

摘要

•The Early Author Profiling task is feasible in social media documents.•Very small amounts of textual evidence can be modeled from early stages.•The proposed meta-words capture discriminative information of target profiles.•The proposed multi-resolution representation boosted the early recognition.

论文关键词：Early text classification,Author profiling,Social media analysis,Text mining

论文评审过程：Received 29 March 2019, Revised 15 August 2019, Accepted 31 August 2019, Available online 2 September 2019, Version of Record 8 September 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.112909