Distributed Spatio-Textual Stream Data Query Processing System

With the development of global positioning technology and location-based services,a large amount of text data is accompanied by spatial location information.As the fusion of geographic data and text data becomes more and more common,the related research of spatio-textual data has become one of the hot spots in the field of spatio-temporal data mining.However,the use value of data diminishes with time,and the valuable information obtained in time has an important impact on the user's decision-making.Therefore,more and more researchers have begun to study real-time processing algorithms for spatio-textual data.Continuous query of spatio-textual data is the main research content of this article.Different from the traditional query,the continuous query starts from the registration in the system until it expires or is deleted by the user.It continuously checks the objects entering the system and finds all the objects that meet the query constraints.However,the existing spatio-textual data stream processing system has limited processing efficiency for continuous queries,does not make full use of the skew characteristics in the data,and lacks the ability to adjust the text information changes in the stream data.In response to the above problems,the distributed spatial range keyword continuous query algorithm was researched,and the main research contents are as follows.(1)An efficient distributed index structure and query algorithm is proposed,which effectively utilizes the spatial and text attributes of the data,minimizes the total load cost while ensuring the load balance of the system,and improves the query efficiency of the system.In order to efficiently use computing resources.(2)The hot and cold keywords are used to divide the data space,and a new load cost calculation model is proposed.While considering the load balance of partitions,the keywords between partitions are not similar as much as possible,and more partitions intersecting with the query but not intersecting with the keywords are divided,so as to reduce the number of query registration and further reduce the system load.(3)As spatio-textual data and continuous query requests continue to enter the system,the load situation of the system will gradually change.In order to ensure the stability of the system,a multi-level dynamic load adjustment strategy is designed to cope with the dynamic changes of the load in the data stream processing environment.The system periodically detects the load conditions of each working node and makes corresponding adaptive adjustments to abnormal load conditions.(4)The performance of the designed system was verified experimentally on different data sets.The experimental results show that the designed distributed spatial text data stream query processing system has high efficiency and reliability.