Analysis And Research In The Feature Of The Spatio-Temporal Object

With the rapid development of computer technology and the increase number of the space-time collection devices.The spatial and temporal data is increasing at an alarming rate.automatic and semi-automatic pattern analysis becomes more essential.Obtain interest-ing, valuable, meaningful patterns from a large number of spatial and temporal data become challenges to spatio-temporal data analysis. Cascading spatio-temporal pattern represents subsets of different object-types whose instances are co-located together in a reasonable time slots.and each pattern has a direction judged by the instance’s time slot.In this paper, our arm is to mine Cascading spatio-temporal patterns from spatiotem-poral data set. According to the existing Cascading spatio-temporal pattern mining analisis. using a cascade mining algorithm based on time slot to analysis and research in the feature of the Spatio-Temporal object. The main research work are as follows:(1)Improve the calculation efficiency of mining Cascading spatio-temporal patterns. In order to improve the calculation efficiency of mining Cascading spatio-temporal patterns and reduce the Computation by reduce the number of the candidate set. we use the filter(UB) to prune mode whose CPI does not meet the given Threshold.(2)We use time slot to deter-mine cascading relationship between different event instances. And assume that instances which happened in the same time slot does not have the cascading relationship. This assume simplifies the time complexity of algorithm. Give the concept of the first time slot and the last time slot. Give the concept of the Imbps and Imps and use it to reduce the number of the Cascade candidate.(3) have a study on Spatial-temporal Co-occurrence Pattern, give naive and fast-PPCop algorithm of the spatio-temporal partial co-located, and made a comparison between the two algorithm, and also made the algorithm analysis and verification.Finally, an experiment was given to verify the positive identification of the study of the spatial-temporal cascade algorithm. We selected and refined the dataset of HangZhou crime data, and the thetic data was plus.