Synthesis of Classification Models and Review in the Field of Machine Learning

Synthesis of Classification Models and Review in the Field of Machine Learning

Venkatram Kari (VIT University, India) and Geetha Mary Amalanathan (VIT University, India)
Copyright: © 2019 |Pages: 34
DOI: 10.4018/978-1-5225-7796-6.ch002

Abstract

Classification method is an important technique used in machine learning for predictive analytics. Classification enables business to predict future trends and behaviors of an enterprise with the help of their past data. Classification is a supervised learning model, which is built in twostep process, first building the classification model and second predicting the outcome for unknown data. This chapter describes various classification models by learning mechanisms and categorizes them into different statistical, probabilistic, and heuristic methods, and explains them with example dataset. It also compares these models and their efficiencies with model evaluation techniques and briefs some blended classification models. The goal of this chapter is to provide a comprehensive review of different classification techniques and give a quick refresher on classification models in big data analytics. The comparison of various classification models helps the readers to quickly decide which classification model to choose for the given business scenario.
Chapter Preview
Top

Background

A typical classification model is shown in the following diagram Figure 1, with two processes 1) Model construction for known dataset with predefined class labels. Build the model with a training dataset. Validating the model on the validation dataset. 2) Classify the unknown data.

Complete Chapter List

Search this Book:
Reset