期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2014
卷号：2
期号：11
出版社：S&S Publications
摘要：Today, various applications require the ability to monitor a continuous stream of fine-grained data forthe occurrence of certain high-level activities. A number of computerized systems—including ATM networks, webservers, and intrusion detection systems—systematically track every atomic action we perform, thus generatingmassive streams of times tamped observation data, possibly from multiple concurrent activities. In this paper, weaddress the problem of efficiently detecting occurrences of high-level activities from such interleaved data streams. Asolution to this important problem would greatly benefit a broad range of applications, including fraud detection, videosurveillance, and cyber security. We define algorithms for insertion and bulk insertion into the tMAGIC index andshow that this can be efficiently accomplished. We also define algorithms to solve two problems: the “evidence”problem that tries to find all occurrences of an activity (with probability over a threshold) within a given sequence ofobservations, and the “identification” problem that tries to find the activity that best matches a sequence ofobservations. We introduce complexity reducing restrictions and pruning strategies to make the problem—which isintrinsically exponential—linear to the number of observations. Our experiments confirm that tMAGIC has time andspace complexity linear to the size of the input, and can efficiently retrieve instances of the monitored activities.
关键词：Activity detection; indexing; stochastic automata; times stamped data; Data mining applications
