An Analytical Study on Machine Learning Techniques

An Analytical Study on Machine Learning Techniques

Law Kumar Singh, Pooja, Hitendra Garg, Munish Khanna, Robin Singh Bhadoria
Copyright: © 2021 |Pages: 21
DOI: 10.4018/978-1-7998-5876-8.ch007
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

The last few months have produced a remarkable expansion in research and deep study in the field of machine learning. Machine learning is a technique in which the set of the methods are used by the computers to make prediction, improve prediction and behavior prediction based on dataset. The learning techniques can be classified as supervised and unsupervised learning. The focus is on supervised machine learning that covers all the predictions problem for which we had the dataset in which the outcome is already known. Some of the algorithm like naive bayes, linear regression, SVM, k-nearest neighbor, especially neural network have gain growth in this area. The classifiers of machine learning are completely unconstrained with the assumptions of statistical and for that they are adapted by complex data. The authors have demonstrated the application of machine learning techniques and its ethical issues.
Chapter Preview
Top

1. Introduction

Machine learning is the branch of artificial intelligence that analyzes and responds to the data on the basis of similar pattern collection called clusters, which helps in analyzing and responding to the data. It is like a blank human mind that understands any image by its physical structure or the data it has from its past experience (Smola and Vishwanathan, 2008).

Key features of machine learning

  • •

    It converts the data into information where the data can be in the form of image, videos, texts, CSV etc

  • •

    It is a problem-solving tool.

    • ex: n-queens, 8-queens, traveling salesman, etc

  • •

    It is the combination of computer science, engineering, and statistics, where computer science means it performs the programming work; engineering means applying different types of applications and statistics means the application of mathematical computations.

  • •

    It interprets data and acts on it, means it takes the data and analyzes that data, learns that data, and based on that learning, it performs actions.

  • •

    It optimizes the performance criteria using past information.

For example, in the finance sector, with the help of past data and algorithms given to a machine, it detects the difference between legal and fraudulent activities, if in case any illegal transaction takes place the machine can detect it easily.

In the e-commerce sector: by calculating the previous record, a machine gets to know customer's interest and sellers conditions, and by using a mathematical algorithm, it gives results having maximum profits for sellers.

Top

2. Types Of Machine Learning

Figure 1 shows the detailed block diagram of the type of Machine Learning.

  • •

    Supervised Learning

  • •

    Unsupervised Learning

  • •

    Semi-Supervised Learning

  • •

    Reinforcement Learning

Figure 1.

Types of machine learning

978-1-7998-5876-8.ch007.f01

Complete Chapter List

Search this Book:
Reset