Mining tree structured data on multicore systems英文文献.pdfVIP

Mining Tree-Structured Data on Multicore Systems Shirish Tatikonda Srinivasan Parthasarathy The Ohio State University The Ohio State University Columbus, OH 43210 Columbus, OH 43210 tatikond@ srini@ ABSTRACT studying glycan molecules which are responsible for many cellular Mining frequent subtrees in a database of rooted and labeled trees is processes [13], and phylogenies which denote evolutionary rela- an important problem in many domains, ranging from phylogenetic tionships among different organisms [25, 42]. analysis to biochemistry and from linguistic parsing to XML data In case of web log mining, the visitor accesses to a website can analysis. In this work we revisit this problem and develop an archi- be modeled (with some approximations) as trees [42]. Frequent patterns extracted from these trees can help in making recommen- tecture conscious solution targeting emerging multicore systems. Specifically we identify a sequence of memory related optimiza- dations, in web personalization, and in better organization of web tions that significantly improve the spatial and temporal locality pages [26]. Frequent tree mining is also found to be useful in an- of a state-of-the-art sequential algorithm – alleviating the effects alyzing XML repositories [44], in designing caching policies for of memory latency. Additionally,