摘要
[目的/意义]根据用户网络查询重构行为预测查询重构策略,可以为用户提供查询帮助,提高用户网络查询交互体验。[方法/过程]以预测模型构建过程为基础对最新研究成果进行梳理,包括查询重构策略划分、查询重构行为特征选取、查询重构情境影响因素,以及预测模型构建方法,揭示研究进展与存在问题,最后提出研究展望。[结果/结论]对查询重构策略预测研究还有较大探索空间,展望可为后续相关研究提供一定参考。
[Purpose/significance]In order to improve the user’s experience of web query,it is very important to predict the strategy of query reformulation based on the query behaviors.[Method/process]Based on the construction process of prediction model,the latest research results are sorted out,including classification of query reformulation strategy,selection of behavior characteristics of query reformulation,influencing factors of query reformulation,and construction methods of prediction model.The research progress and existing problems are revealed.Finally,the research prospects are put forward.[Result/conclusion]There is still much room for further research on query reformulation strategy prediction,prospects can provide some reference for the follow-up related research.
出处
《情报理论与实践》
CSSCI
北大核心
2020年第2期171-176,170,共7页
Information Studies:Theory & Application
基金
国家自然科学基金项目“嵌入式知识服务驱动下的领域多维知识库构建”(项目编号:71573102)
吉林大学基本科研业务费项目“查询重构有效性预测模型构建”(项目编号:2017QY005)的研究成果之一
关键词
查询重构策略
查询重构行为
影响因素
预测模型
综述
query reformulation strategy
query reformulation behavior
influence factor
prediction model
review
