Group Pattern Mining Algorithm Of Moving Object On Uncertain Data

With the rapid development of mobile phones,GPS,LBS application technology,the formation of the massive mobile data,is closely related to these mobile data and human itself,so the trajectory data analysis has a important significance for research and commercial.In addition,due to the equipment acquisition accuracy,network transmission delay and other reasons,the uncertain data exists in reality,so for uncertain data mining algorithm has become a hot topic of current research.Due to the traffic flow analysis,mobile electronic commerce,based on the location of the service and wildlife research significance,this paper addresses the uncertain data set of patterns mining algorithm,this pattern mining for understanding of the uncertainty trajectory data is very important.Trajectory data set of patterns mining content mainly includes two aspects:one is clustering analysis.It is in a constantly moving object clustering problem;second is the set of patterns of mining,mining its is in a moment all belong to the same clustering objects.The proposed algorithm needs to solve the following problems:(1)to select the appropriate clustering algorithm for group pattern mining provide correct uncertain data as input;(2)how could the world model to join the group model oriented to determine the data mining algorithm;(3)the search space of the algorithm is very huge,how to carry out cutting optimization;(4)The original algorithm and optimization experiments were compared,the two algorithms uptime and mining results were compared to verify whether the algorithm can improve the efficiency of the optimization algorithm to ensure mining results simultaneously.The clustering algorithm as group pattern mining preprocess operation for group pattern mining of uncertain data,clustering algorithms including the DBSCAN and c-means clustering and so on,we use the fuzzy c-means clustering to determine the trajectory data is converted into uncertain data clustering,and then put forward for uncertain data set of patterns mining algorithm to get the final result.According to the train of thought of the research the data track group pattern mining applied to uncertain data,the improved for ObjectGrowth algorithm,to enable it to the same mining operations on uncertain data.And a new optimization method based on the existing cutting rules is proposed to optimize the search.