Font Size: a A A

Social-aware Top-k Spatial Keywords Search

Posted on:2017-02-15Degree:MasterType:Thesis
Country:ChinaCandidate:Q J CuiFull Text:PDF
GTID:2308330503982193Subject:Computer technology
Abstract/Summary:PDF Full Text Request
With the popularity of portable mobile devices, location-based geographic information services are more and more important in people’s lives. With the users’ s needs, Spatial keyword queries cannot return satisfactory results to users. Meanwhile, a large number of data about geographical and social attributes generate by social network services bring great challenges to study the efficience of query. In the basis of above problems, Social-Aware Top-k Spatial Keywords Search is proposed by combing social information with the spatial keyword search.Firstly, because of the user’s choice often affected by the recommendation of friends, the use of location-based services generated by the user registration records and social data, a new query algorithm is proposed, which called social-text-aware top-k location query. In this algorithm, social information and user’s preferences is applied to the spatial keyword search and k objects interested by users are returned. The selected k objects depend on three aspects: location proximity, textual similarity and social similarity.The sum of these three similaritits are treated as a measure to return k the highest score of k objects. In the query process.In order to reduce the query response time, three pruning strategies that including distance, text and social pruning strategy are proposed.Secondly, in view of the traditional preference query only considers the score of the auxiliary service facilities, and the score of the target facilities service is neglected, unsatisfactory of the result. A new query called social-text-aware top-k preferences query is developed. social-text-aware top-k preferences query integrate the user’s friends and friend’s checkins data that were used to into preferences query. A preference query retrieves the k features objects with the highest scores. The selected k objects depend on the score of auxiliary features, features and social relations score of the service facilities.To speed up the query speed, 2-way termination condition are proposed, which cut objects from two aspects successively.Finally, empirical analysis on data set verifies the validity of our approaches.
Keywords/Search Tags:social relevance, preference query, spatial keyword search, SNIR-Tree
PDF Full Text Request
Related items