EFFECTIVE MODELLING FOR PREDICTIVE ANALYTICS IN DATA SCIENCE

Abstract

Predictive analytics includes many statistical and other empirical methods that create various data predictions as well as different methods for assessing predictive power. Predictive analytics not only helps in creating practically useful models but also plays an important role in building new theory for further study and research. Today, the use of available data to extract inferences and predictions by using predictive analytics has grown in the industry from being a small department in large companies to being an active component in most mid to large sized organizations. This paper addresses to reduce a particularly large gap of, the near- absence of empirical or factual predictive analytics in the mainstream research going on in this field by analyzing the issues faced in predictive modelling by the empirical determination of data with its experimental facts for latency pattern.