Fuzzy SVM Clustering Using Plane Sweep Algorithm in Big Data Mining

The era of Big Data is leading to an enormous growth of digital information. Text mining is akin in nature to data mining. Text mining is the practice of examining and discovering big amounts of unstructured text data assisted by software that can classify patterns, topics, concepts, keywords and other attributes in the data. Powerful technique for data classification is Support Vector Machine (SVM). SVM makes a decision surface by creating an optimal separating hyper-plane, to divide the data points of different kinds in the vector space. Clustering the text takes a special place since it is reliable and easy to configure. To cluster big datasets and to shrink the computational time, Hadoop distributed platform, MapReduce programming standard is used by issuing clustering job across various computing nodes or handling Big Data MapReduce is proven state of art technology. In Big Text Data analysis MapReduce provides a faster implementation and is a powerful tool. In the event of a big data like server having a difficulty in providing the services furthermore, additional servers in the cluster can take the load. A powerful approach for solving difficulties including geometric objects in a plane known as Plane Sweep algorithm. The key benefit of using plane sweep algorithm is that it clusters more related data into a cluster hence retrieval will be extra fast and accurate. The Evaluation result shows that the proposed approach provided better result when compared with other approaches.