期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2017
卷号：95
期号：8
出版社：Journal of Theoretical and Applied
摘要：Many business organizations exploring their big data resources using Hadoop distributed file system due to its capability in offering fast processing, easy scalability and higher extend reliability on low cost commodity hardware. The efficiency of Hadoop distributed file system is the result of using Mapreduce frame work, which is leading research communities to scale conventional data mining techniques to fit on Mapreduce framework. So far in the research we can found proposals of data mining on Mapreduce frame work concentrating only on scalability factor, but at the same time they are ignoring two important factors which can affect the efficiency, those are reducing number of scans to database and lacking of intuitiveness in obtained data mining results. In order to overcome such short comes and considering importance of classification learning techniques in real time environment, we propose an fuzzy associative classifier learning model using Mapreduce framework which take advantage of Tid-list representation to extract the total classifier with in single scan to database and provides intuitiveness using data driven fuzzy clusters. The experimental results show the proposed model successfully enhanced the fast and intuitive efficiencies of classification techniques for big data analytics without compromising the accuracy.
