Distance measures in author profiling
作者：
Highlights：
• We verify whether 24 distance measures used frequently respect some useful theoretical properties.
• We verify empirically the effectiveness of 24 distance measures based on 13 test collections used in author profiling tasks.
• We measure the effectiveness impact of changing the text genre between the learning and testing phase in the context of author profiling problems.
摘要
•We verify whether 24 distance measures used frequently respect some useful theoretical properties.•We verify empirically the effectiveness of 24 distance measures based on 13 test collections used in author profiling tasks.•We measure the effectiveness impact of changing the text genre between the learning and testing phase in the context of author profiling problems.
论文关键词：Distance measure,Author profiling,Pan-clef,Text categorization
论文评审过程：Received 29 November 2016, Revised 4 April 2017, Accepted 19 April 2017, Available online 4 May 2017, Version of Record 4 May 2017.
论文官网地址：https://doi.org/10.1016/j.ipm.2017.04.004
