New query suggestion framework and algorithms: A case study for an educational search engine
作者:
Highlights:
• Query suggestion (QS) problem is reduced to “comparison of queries” problem.
• A modular and practical framework is suggested for development of QS algorithms.
• Breadth First Search graph traversal method is better for query-log traversing.
• Combining QS algorithms improves the performance.
• Suggested new QS algorithms achieved 66–90% performance increase.
摘要
•Query suggestion (QS) problem is reduced to “comparison of queries” problem.•A modular and practical framework is suggested for development of QS algorithms.•Breadth First Search graph traversal method is better for query-log traversing.•Combining QS algorithms improves the performance.•Suggested new QS algorithms achieved 66–90% performance increase.
论文关键词:Query suggestion,Framework,Educational search engine,Query recommendation
论文评审过程:Received 2 March 2015, Revised 16 November 2015, Accepted 1 February 2016, Available online 15 March 2016, Version of Record 22 July 2016.
论文官网地址:https://doi.org/10.1016/j.ipm.2016.02.001