A Survey of Network Faults Classification Using Clustering Techniques

Abstract: A massive data mining and knowledge discovery present a huge challenge with the volume of data growing at an unprecented approximation value. MapReduce has been implemented in manage many large-scale computation. The recently introduced MapReduce technique has received much consideration from both scientific community and industry for its applicability in big data analysis. The effective computation of approximation is essential improving the performance of data mining and other related task. For the purpose of data mining for massive data, parallel computing modes and algorithms are typical methods in these research fields. To mine knowledge from big data, we present consequently algorithm corresponding to the MapReduce based on roughest theory are put forward to deal with the massive data, in this paper comprehensive to evaluate the performances on the large data sets show that the proposed demonstrated can effectively process of big data.

Keywords: Data mining, MapReduce, HDFS (Hadoop Distributed File System), Rough sets