计算机应用研究2016，Vol.33Issue(3)：657-660,4.DOI:10.3969/j.issn.1001-3695.2016.03.004
基于查询热度和实体识别的查询推荐
Query recommendation based on query’s hot degree and entity recognition
摘要
Abstract
Query recommendation has become an important way to improve the user search’s experience and the quality of service of the search engine.Improving the quality of the recommended query and user’s satisfaction is particularly urgent. Previous studies ignored the importance of overall log information and the named entity of the search log.By assessing the heat degree of the clustered query and extracting the named entities in the query string,this paper proposed a new query recommen-dation way by fusing query’s hot information and named entity to the similarity calculation formula.This method’s average value of the satisfaction is higher than the latest three methods’value.It indicates this method’s validity.This method took advantage of the recognized named entity and considered the hot degree of the recommended query string in the global search log.It improves the overall quality of the recommended query string.But this method is limited to the accuracy of the extracted features,depends on the further abundant and optimization.关键词
聚类/特征提取/热度/命名实体/模板权重/查询推荐Key words
clustering/feature extraction/hot degree/named entity/template weight/query recommendation分类
信息技术与安全科学引用本文复制引用
任育伟,吕学强,李卓,徐丽萍..基于查询热度和实体识别的查询推荐[J].计算机应用研究,2016,33(3):657-660,4.基金项目
国家自然科学基金资助项目（61271304）；北京市教委科技发展计划重点项目暨北京市自然科学基金 B 类重点资助项目 （）
