A Map Reduce Based Support Vector Machine for Big Data Classification in Hadoop Bigdata

Now a days people are enjoying the world of data because size and amount of the data has tremendously increased which acts like an invitation to Big data. But some of the classifier techniques like Support Vector Machine (SVM) is not able to handle the huge amount of data due to it's excessive memory requirement and unreasonable complexity in algorithm tough it is one of the most popularly used classifier in machine learning field. Hence a new technique comes into picture which performs parallel algorithm in a efficient way to work data having large scale called as PSVM. In this paper we are going to discuss a PSVM model for risk analysis which is based on map-reduce, and can easily handle a huge amount of data in a distributed manner.