A Review Paper on Machine Learning in Pattern Recognition

Abstract

Supervised or unsupervised classification is the main goal of pattern recognition. The statistical approach is the most popular approach that prevails among the many frameworks in which pattern recognition is initially formulated. Recently, more attention has been paid to neural network techniques and methods derived from statistical learning theory. This requires attention to the design of the sensor system. There are various issues related to the design of identification systems. They are the definition of pattern classes, the detection and extraction environment, representation and feature selection, cluster analysis, classifier design, learning and selection of training and test samples. There is no solution to the general problem of complex pattern recognition associated with arbitrary patterns. Data mining, web search, and multimedia retrieval have several emerging applications that require appropriate and effective tuning techniques. The main goal of this article is to provide a detailed description of different methods that can be used in different stages of a pattern recognition system. The purpose of this paper is also to explore research topics in application that can be highlighted in this challenging field.