Comparative Study of Big Data Predictive Analytics Frameworks

Abstract

Big data in terms is the huge volume of data that is difficult to be processed, handled and managed. The term big data is an emerging trend where a number of researches and data scientists are carried out for data analytics. One of the most interesting thing in big data analytics is about the predicting for future by using the data. Predictive analytics is the use of data and machine learning (ML) techniques to identify the future outcomes based on historical data. With predictive analytics a company can meaningfully leverage that business data to diagnose and solve business problems. But choosing a proper predictive analytics framework need to be consider many things, and it is also vital for every data analysis project. The best and the simplest practical way is to compare the response time of each framework. In this paper, we will investigate and compare two big data predictive analytics frameworks, Apache Mahout and Spark MLlib, from performance point of view. This comparative study will make things easier to the researcher and data scientists in the selection of big data analytics frameworks according to their analytics areas.