Research And Implementation Of Spatial Text Data Query Processing Technology

With the extensive application of location technology and the flourishing of the Internet,a large amount of textual data is fused with spatial location information.How to quickly get objects that users are interested in in the large amount of spatial text data has has become a research hotspot in recent years.The existing spatial keyword query processing techniques have the following deficiencies and limitations: firstly,it is impossible to return the objects that satisfy multiple users under the spatial keyword query proposed by multiple users.Secondly,when the spatial keyword range query with user preference constraints is processed,the existing index does not take into account the user's preference attribute,which leads to low query efficiency.Finally,the response speed of query processing is not guaranteed when processing massive spatial text data.In this paper,in view of the shortcomings of the existing algorithms,we study the spatial text data query processing technology in the actual scene,the main research contents are as follows:(1)Traditional spatial keyword queries are presented by a single user,which contains a query location and a set of query keywords.Collaborative Spatial keyword top-k query(TKCSKQ)returns top-k objects that are close to multiple query positions and have high relevancy between object's text and multi-group query keywords.In view of the phenomenon of repeated and near keywords in multigroup query keywords in TKCSKQ,a keyword relevance calculation formula based on query keyword weight is designed.SKNIR-tree index is proposed supporting near keywords matching by extending IR-tree index.Based on the SKNIR-tree,collaborative spatial keyword top-k query algorithm(TKCSK)is proposed,pruning through maintaining a priority queue and calculating the minimum spatial textual correlation of each node with the query,to quickly identify the desired objects.The experimental results show that The TKCSK algorithm has excellent performance compared with the existing algorithm.(2)The existing indexes for spatial keyword range query do not consider user preferences,resulting in poor pruning performance and low query efficiency.In order to solve this problem,a hybrid index called BRPQ(Boolean Range with Preferences Query index)is proposed to support user preferences,spatial location and keywords collaborative pruning.We also propose an efficient query-processing algorithm for spatial keywords range query with user preferences constraint.Experimental results show that BRPQ outperforms the existing indexes in terms of building time and query processing efficiency.(3)For massive spatial text data,the traditional index structure and query algorithm have single machine limitations,and cannot meet users' needs in terms of storage capacity and processing speed.To solve this problem,a spatial keyword range query mechanism based on HBase is proposed.Designing spatial region partitioning strategy and rowkey index spatial information and text information at the same time,to support parallel processing and extension of spatial keyword range query.The experimental results show that the algorithm proposed in this paper is superior to the traditional algorithm.