期刊名称：International Journal of Computer Science and Network
印刷版ISSN：2277-5420
出版年度：2016
卷号：5
期号：2
页码：365-368
出版社：IJCSN publisher
摘要：We propose a system for data mining dynamicXML document using a GPU. As a GPU is capable ofprocessing in parallel in real time, it promises on increasingthe speed of data discovery process. Nvidia has provided itsCUDA platform to utilize its graphics processor forimplementing non-graphical algorithms on it. The WWW isstructured over the Extensive Markup Language (XML) andhence provides access to the data structured over the WWW.The main objective is to find association rules by extractingitems on XML nodes into the XML documents. The use ofGPU comes in picture in the pre-processing phase andsorting phase when the item set is being shortened over theoccurrences of the items, as this process takes a lot of time theuse of a GPU promises a very significant decrease in the preprocessingtime.
关键词：GPU; Data Mining; XML; CUDA; Parallel;Processing.
