A top-k query results ranking approach based on contextual preferences for web database

作者：Meng Xiang Fu^*; Ma Zong Min; Li Xin; Zhang Xiao Yan; Wang Xing

来源：Chinese Journal of Computers, 2014, 37(9): 1986-1998.

DOI：10.3724/SP.J.1016.2014.01986

• Clustering
• Contextual preference
• Preference modeling
• Rule mining algorithms
• Top-k query
• Top-k rankings
• User's needs
• Web database

To deal with the problem of too many results returned from a Web database, this paper proposes a contextual preference-based top-k query results ranking approach. A contextual preference model with an interest degree, which takes the form of i1≻i2, d|X, meaning that item i1 is preferred to item i2 with an interest degree d in the context of X, is proposed. The contextual preferences can be obtained from query history by using the association rule mining algorithm. Based on the contextual preferences, a top-k query results ranking method is proposed and the corresponding tuple's order creating, clustering and top-k ranking algorithms are presented. The results of experiments demonstrated that the contextual preference model has a strong expression ability of preferences, the top-k ranking method can meet the user's needs and preferences closely and it has high performance as well.

• 出版日期2014
• 单位东北大学