期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2016
卷号：5
期号：10
页码：2482-2487
出版社：Shri Pannalal Research Institute of Technolgy
摘要：With exponentially increasing electronic data dayby day, Big Data is gaining attention for solving faster accessand summarization problems. However, this huge amount ofdata with heterogeneous formats compelled us to renovate ourtraditional use of learning algorithms and ponder about newtechniques which are challenging and complex. To solveproblem of big data, we propose a linguistic fuzzy rule basedclassification system, which mainly consist of two methods viz.FuzzyReducerMax and FuzzyReducerAve. As name fuzzysuggest vague and uncertain in the similar way it is dealingwith uncertainty that is essential to the diversity andauthenticity of big data and because of the procedure oflinguistic fuzzy rules it is capable to render a recognizable andoperational classification model. This process is established onthe MapReduce framework, which are very popular andfrequently used to handle big data by Hadoop framework. Theperformance measure is done on these methods by using aData set of networking attack logs. The result shows itscapability to provide accuracy on classification with both theapproaches and runtime analysis which shows its speedimprovement.
关键词：Big Data;Classification;Hadoop;Map Reduce;NLP; Parts of Speech Tagging
