The Research On The Related Problems Of Association Rule Mining Over Data Streams

Recently, large amounts of data are accumulated in the form of data streams, such as web data and transaction data. Different from in traditional static databases, data streams are continous,unbounded,usually coming with high speed and having a distribution that often change with time.Traditional mining algorithms are difficult to cope with data stream due to its characteristic,such as consecution, disorder and real-time.This poses many new challenges for the knowledge discovery of mining data streams, and mining frequent patterns over data streams has become current research difficulty and hotspot.The paper mainly study some of data stream mining problems.The detail research achievements are listed as follows:Firstly, mining technologies over data streams and their characteristics is introduced.Then the basic conceptions and key problems is introduced. At last, the thesis analyes several typical algorithms of mining in data streams.Secondly, a new PCFI algorithm based on sliding window for mining frequent cloesd patterns in data stream is proposed. A sliding window is divided into several basic windows and the basic window is served as an updating unit. A compact tree PCFI-tree is used to mine frequent closed patterns in the basic window and to maintain all the frequent patterns. The obsolete and infrequent items are deleted. The experimental results indicate that PCFI algorithm performs efficiently.Finally, a new algorithm (MMFIDS) for mining maximal frequent patterns based on damped transactions in data streams was proposed. The title-time window was adopted in the algorithm. In addition, the methods which were the bit objects for data expression,bit frequent patterns'tree( IFP-tree) and storing patterns'tree(PTTW-tree)were used to store and handle the data. It could mine maximal frequent patterns in data streams that were made up of new or old transactions. The experimental result verifies the efficiency of the algorithm.