Study On Top-k Collection Spatial Keyword Approximation Query Method

With the continuous development of GPS and mobile network technology and the popularization and application of intelligent devices,a large number of spatio-textual objects containing Location information and text information appear on the Web,which makes Location Based Service(LBS)widely used.In recent years,the scale of spatio-textual data with location and text information has been increasing rapidly.Spatial keyword query technology based on spatio-textual objects is an important supporting technology of location-based service system.Spatial keyword query has become a research hotspot in spatial database and information retrieval.Collective spatial keyword query is a kind of spatial keyword query.The association relationships between spatial objects in a group are not considered while there are usually closely social correlations between spatial objects in the same group.The greater the association relationships between spatio-textual objects,the more likely the user would visit these objects at the same time,and the results are more in line with the user’s need.However,such kind of methods does not consider the relationships(such as social correlations,textual similarity)between spatial objects in the group,which lead to the objects covering all query keywords but not closely related to be divided into a group,which can not meet the actual needs of users.To deal with this problem,this paper proposes a top-k collective spatial keyword approximate query method,and has achieved the following innovative research results:Considering the association relationship between spatio-textual objects in the spatio-textual object group,an association relationship evaluation method between spatial objects based on association rules is proposed.Studying an accurate query method of collective space keywords based on association relationship.And we design a scoring function which combines the location distances and social relationships of spatio-textual objects within a group.Using a strategy based on local neighborhoods to reduce the number of candidate spatio-textual object groups.And then VP-Tree based pruning strategy is proposed for quickly searching the local neighborhood of spatio-textual objects,so as to speed up the query speed.And we use this to realize the Top-k collective spatial keyword approximate query method.Through experiments on real data sets,the results show that the query results obtained by the scoring function proposed according to the spatio-textual object association relationship are more reasonable and closer to the needs of users.The proposed pruning strategy has high execution efficiency.The above research results can be used in spatial text data query,location-based services,urban planning and other application fields.This thesis has 20 figures,17 tables and 77 references.