Active tag recommendation for interactive entity search: Interaction effectiveness and retrieval performance
作者：
Highlights：
• We introduce active tag recommendation; an approach using online reinforcement learning for recommending tags for the user to support the information search process.
• We show that active tag recommendation improves the ranking of search results.
• We show that active tag recommendation increases the amount of interaction and effectiveness of interactions to rank selected information.
• We show that active tag recommendation reduces, but does not substitute typed-query interaction.
• We show that active tag recommendation does not compromise task execution time.
摘要
•We introduce active tag recommendation; an approach using online reinforcement learning for recommending tags for the user to support the information search process.•We show that active tag recommendation improves the ranking of search results.•We show that active tag recommendation increases the amount of interaction and effectiveness of interactions to rank selected information.•We show that active tag recommendation reduces, but does not substitute typed-query interaction.•We show that active tag recommendation does not compromise task execution time.
论文关键词：Tag recommendation,Active learning,Information retrieval,Search user interfaces,User study
论文评审过程：Received 8 October 2021, Revised 26 November 2021, Accepted 23 December 2021, Available online 10 February 2022, Version of Record 10 February 2022.
论文官网地址：https://doi.org/10.1016/j.ipm.2021.102856
