Supervised approaches for explicit search result diversification
作者：
Highlights：
• We leverage supervised learning methods to improve the effectiveness of explicit search result diversification.
• We cast the diversification problem as that of learning a ranking model, based on the coverage of query aspects by each candidate document.
• We learn the importance of query aspects by re-ranking the candidate documents for each aspect and leveraging query performance predictors.
• We cast the diversification problem as a fusion task, namely, the supervised merging of rankings per query aspect.
摘要
•We leverage supervised learning methods to improve the effectiveness of explicit search result diversification.•We cast the diversification problem as that of learning a ranking model, based on the coverage of query aspects by each candidate document.•We learn the importance of query aspects by re-ranking the candidate documents for each aspect and leveraging query performance predictors.•We cast the diversification problem as a fusion task, namely, the supervised merging of rankings per query aspect.
论文关键词：Explicit diversification,Supervised learning,Query performance predictors,Aspect importance
论文评审过程：Received 28 March 2020, Revised 14 June 2020, Accepted 5 July 2020, Available online 30 July 2020, Version of Record 30 July 2020.
论文官网地址：https://doi.org/10.1016/j.ipm.2020.102356
