Research On Data Stream Reverse K Nearest Neighbors Outlier Mining Algorithm Based On X~* Tree

Data stream mining technology is an effective approach to resolve the problems of abundant data and scanty information in modern Web Times.It currently is the research frontier within the information science field.It has been deemed as a field that has broad prospect of application in database research.In many data mining algorithms,outlier mining is an important matter in data mining,in which process that building high efficient index structure is a key problem in mining data stream outlier.An urgent requestion is putting forward for the data stream outlier mining algorithm which needs few times of scanning database and less request of memory.On the base of analysis of the already existing data stream outlier mining algorithms,we research the problem of outlier mining in the following aspects.Firstly,this paper researches the X*tree index structure which supports knn searching.Because of the inefficiency of the time and memory consuming of the X-tree index structure when performing knn query,this paper introduces a simple clustering algorithm for the splitnode algorithm which supports knn query better.The new index structure is kown as X*tree which consumes less memory than X-tree.SplitHistory infomation is then omitted.Secondly,this paper researches data stream outlier mining algorithm which is based on Reverse k nearest neighbours.When we analyze the known algorithms,we find that the algorithm cannot deal with the concept drifting problem and they need multi-scan of the dataset.So,this paper introduces the SODRNN algorithm,which needs only one pass of scan for the current sliding window.Finally,the empirical study verify the feasibility and effectiveness of X*tree index structure which supports knn searching and the SODRNN algorithm in this paper.