深圳大学学报理工版

查询推荐是指根据用户的输入提供若干替代的查询,用户使用推荐的查询去检索,得到更多符合需求的信息.利用基于位置的关键词查询推荐所提供的替代关键词能够检索到在用户查询位置附近的信息.用户提交的关键词常是多义词且含有各自的背景偏好,采用具有个性化的推荐查询则能检索到符合用户偏好的信息.为同时满足空间位置邻近和个性化需求,提出一种基于位置的个性化关键词查询推荐方法,使推荐查询的关键词能够检索到位于用户附近且符合其偏好的信息.用关键词-文档二部图表示不同关键词查询之间的语义相似性,采用动态边权重调整策略,建立与关键词相关的文档和用户当前位置的空间关系,使用分类向量模型表示用户的兴趣爱好,应用带重启的随机漫步模型,得到与用户输入的关键词具有较高相似度的其他关键词.在AOL真实数据集上的测试结果表明,该方法为用户推荐的关键词不仅可以满足用户的信息需求,还可以检索到用户位置附近符合其偏好的文档.

The query recommendation provides several alternative queries based on the input query. By using the recommended queries, the users may retrieve more relevant information. Location-aware keyword query recommendation aims for suggesting queries which are able to retrieve the relevant information close to the user's location. When the submitted queries are ambiguous and have various background preferences, the personalized recommendation queries can retrieve information that meets users' preferences. This paper studies a new method of query recommendation, i.e., the location-aware personalized keyword query recommendation. The queries suggested by this approach are able to retrieve nearby relevant information that matches the users' preferences. The proposed method establishes the semantic relationships among keyword queries via a keyword-document bipartite graph. The weights of edges in the keyword-document bipartite graph are dynamically adjusted to represent the spatial proximity of documents. The users' preferences are modeled by the category-based vectors. The random walk with restart model is used to compute recommended queries. This paper develops an efficient algorithm and data structures for the computation of recommendations. The experiments on a real data set AOL demonstrate the effectiveness of the proposed method.

引言
1 基于位置的个性化关键词查询推荐
2 基于位置的个性化关键词查询推荐算法
3 基于真实数据的测试结果及分析
4 结语

图1 K-D图示例<br/>Fig.1 Keyword-document bipartite graph

图1 K-D图示例
Fig.1 Keyword-document bipartite graph

图2 基于位置的二部图权重调整<br/>Fig.2 Weight adjustment based in bipartite graph

图2 基于位置的二部图权重调整
Fig.2 Weight adjustment based in bipartite graph

图3 对历史查询和现有查询的权重分配<br/>Fig.3 Weight allocation to history queries and current query

图3 对历史查询和现有查询的权重分配
Fig.3 Weight allocation to history queries and current query

图4 不同参数设置对算法运行结果<br/>Fig.4 Algorithm running results based on various parameters setting

图4 不同参数设置对算法运行结果
Fig.4 Algorithm running results based on various parameters setting

[1] BAEZA-YATES R, HURTADO C, MENDOZA M. Query recommendation using query logs in search engines[C]// Proceedings of the International Conference on Current Trends in Database Technology. Berlin: Springer-Verlag, 2004: 588-596.
[2] BEEFERMAN D, BERGER A. Agglomerative clustering of a search engine query log[C]// Proceedings of the 6th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, USA: ACM, 2000: 407- 416.
[3] CAO Huanhuan, JIANG Daxin, PEI Jian, et al. Context-aware query suggestion by mining click-through and session data[C]// Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, USA: ACM, 2008: 875- 883.
[4] MIYANISHI T, SAKAI T. Time-aware structured query suggestion[C]// Proceedings of the 36th International ACM SIGIR Conference on Research and Develop in Information Retrieval. Dublin:[s. n.], 2013: 809-812.
[5] CRASWELL N, SZUMMER M. Random walks on the click graph[C]// Proceedings of the 30th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. New York, USA: ACM, 2007: 239-246.
[6] LI Lin, XU Guandong, YANG Zhenglu, et al. An efficient approach to suggesting topically related web queries using hidden topic model[J]. World Wide Web, 2013, 16: 273-297.
[7] SONG Jun, XIAO Jun, WU Fei, et al.Hierarchical contextual attention recurrent neural network for map query suggestion[J]. IEEE Transactions on Knowledge and Data Engineering, 2017, 29(9): 1888-1901.
[8] WU Dingming, CONG Gao, JENSEN C S. A framework for efficient spatial web object retrieval[J]. The Internal Journal on Very Large Data Bases, 2012, 21(6): 797-822.
[9] JÄRVELIN K, KEKÄLÄINEN J. Cumulated gain-based evaluationof IR techniques[J]. ACM Transactions on Information Systems, 2002, 20(4): 422- 446.
[10] TONG Hanghang, FALOUTSOS C, PAN Jiayu. Fast random walk with restart and its applications[C]// Proceedings of the 6th International Conference on Data Mining. Washington D C: IEEE Computer Society, 2006: 613- 622.
[11] BRIN S, PAGE L. The anatomy of a large-scale hypertextual web search engine[J]. Computer Networks ISDN Systems, 1998, 30(1/2/3/4/5/6/7): 107-117.

备注

引言

1 基于位置的个性化关键词查询推荐

2 基于位置的个性化关键词查询推荐算法

3 基于真实数据的测试结果及分析

4 结语

附加材料_增强文件请点击下载

期刊信息

备注

引言