The Studies On Spatial Keyword Personalized Top-k Query Approach

With the rapid development of wireless technology and Internet technology and the popularization of smart mobile devices,location-based services(LBS)have been widely applied,such as Meituan and Baidu map.The use of these applications produces a large amount of spatial-text data that contains both textual and geographic information.Spatial keyword query technology as an effective method for processing space-text data,has received extensive attention from academia and industry.The traditional spatial keyword query technology mostly considers the location proximity and text correlation between the user and the point of interest and ignores the satisfaction of other factors for the user's preference,which cannot be well satisfied User personalized query needs.In order to better meet user needs,we study Top-k space keyword preference query based on influence constraints.Given a set of spatial objects,a set of feature objects,a set of query keywords,the spatial keyword preference query returns the top k best spatial objects,and the existence of one in the neighborhood of each spatial object can make the spatial object better Feature objects that meet user query needs.The query in this paper considers the satisfaction of user preferences of other objects around the target object,so that the query results can better meet the user's personalized query needs.The innovative research contents of this article mainly include:(1)In the query,we consider the satisfaction degree of the facilities around the query object to the user preference,and study the Top-k spatial keyword preference query based on impact constraint.(2)A threshold inverted file algorithm TAIFA is proposed to improve the basic algorithm through the threshold pruning strategy.TAIFA algorithm calculates an upper bound fraction for each R*-tree node.When the upper bound fraction is less than or equal to the current threshold,the node and the sub-trees contained in the node are pruned.(3)In order to further improve query efficiency,we propose a nearest neighbor query algorithm GS-NNA based on greedy strategy.Combining the greedy strategy and the nearest neighbor method,the query results are found by the feature objects with the highest text relevance and meeting the threshold criteria.In summary,we study Top-k spatial keyword preference query based on influence constraints and propose two effective query improvement algorithms.A lot of experiments are carried out on the real data set.By setting different parameters,the effect and performance of the proposed algorithm are evaluated.The experimental results show that the method has better query results and faster execution efficiency.