Research On A Path Query Method For Ordered Spatial Keywords

Spatial keyword query generally returns one or several space service objects for the user under the condition that given a query point and a group of keywords.these objects or the object group can satisfy both the given keywords and the user's space requirements.Most of the research before is based on the Euclidean space,the road network research is not too much.The research on a path query method for ordered spatial keywords is that given a query point and a set of keywords,the result obtained by the final query method is a path that satisfies the conditions defined by the user.The spatial object or spatial object set on the path can not only meet the keyword matching specified by the user and the corresponding specific location requirement of the user,but also meet the order requirement set by the user.The purpose of this paper is to study how to solve the path query for ordered spatial keywords in an efficient way in the road network.First,this paper studies the model representation of road network and the method of dividing road network,and summarizes the spatial index of existing road network,and selects the most suitable index basis for demand.And in order to minimize the need for redundant queries,I did some research on the deletion of technology which is mainly from the text information and spatial information.Next,The PLIR(poly-line IR-tree)global index is proposed by extending the IR tree.The newly constructed PLIR tree can be used to build a LAN query word mapping table,it can quickly retrieve the underlying local area network that contains query keyword.The SCAB +(search cache adjencency B +-tree)tree hierarchical index is proposed,and the query cache table and adjacency file table are used to index the endpoint based on the B + tree.The use of adjacency file table greatly improves the query speed.At the same time,in the border point,it caches the user's query results.The advantage is that querying the same keyword in an endpoint next time no longer needs to expand the query.By using the SCAB+ tree proposed in this paper,it can be more efficient to achieve the query when the query branch extended to the endpoint.Then,based on the index that has been established,the filtering query based on the global index tree is used to query which underlying LAN contains the keywords to be searched.And expand the query based on the hierarchical tree,use short-cuts to skip some LANs that do not contain keywords to reduce redundant queries.The pseudo-code description of query strategy and query algorithm is given on the basis of the index that has been established.Finally,Based on the experimental data and the experimental platform,The validity of the method proposed in this paper is verified by experiments.