Abstract

In the past decade, machine learning techniques have been used for solving several problems with respect to big data. In current, there are several types of Machine Learning (ML) techniques available like supervised, unsupervised and semi-supervised. Similarly, several techniques like classification, Pre-processing, Association rules, Random forest, Decision tree, Support vector machines, etc. available to solve several problems like data imbalance, machine translation, enhancement in robotics, etc. Today’s we need to several basic facts about machine learning techniques to solve many problems like prediction analysis in several applications, for example, in e-healthcare applications (big data: data generated from connected electronically smart devices)/ in other applications (agriculture, e-commerce, defence, etc.). For that, most of the researcher confused and hesitate to discuss/ decide which technique or metric to use in respective applications. How machine leaning techniques differs from Data mining techniques? So this article reviewed several existing work/ papers and presents a remedy for all (those) researchers, which does not solve only doubt from their first stage (selection of techniques or metrics and doubts regarding data mining and machine learning) but also mitigate several issues with respect to machine learning techniques (in compare to deep learning). Hence in summary, this work summarizes with various needful information related to Machine Learning (including Big Data). Following survey on evaluation metrics and some other related factors, this paper showed some future directions at last.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call