Machine Learning and Big Data

9Unsupervised Learning

M. Kumara Swamy¹* and Tejaswi Puligilla^2*

¹ Department of Computer Science & Engineering, CMR Engineering College, Kandlakoya (V), Hyderabad, Telangana, India

² Deloitte Consulting, Madhapur Hyderabad, Telangana, India

Abstract

Machine learning (ML) algorithms train an automated system with an existing data and the trained system is expected identify the class label of the new item. In the ML, the existing data is used to train the system. Hence, the systems are called the supervised learning. In ML, there existing another type of systems called unsupervised learning. In case of unsupervised learning, the training data does not exist to train the automated system. In this chapter, we explain the various ML algorithms/approaches in the area of unsupervised learning. We explain parametric and non-parametric approaches. We also concentrate on Dirichlet process mixture model and X-means algorithms in ML approaches.

Keywords: Machine learning, data sciences, data mining, algorithms, data

This chapter enlighten about various unsupervised learning approaches in machine learning.

• Basic algorithms such as parametric and non-parametric algorithms.
• Dirichlet process mixture model and X-means algorithms in machine learning.

9.1 Introduction

Machine learning (ML) is one of the applications of artificial intelligence (AI) which provides the flexibility to find out from the present data and identify the unknown object from the training. The method of learning starts ...