Machine Learning and Pattern Recognition

Abstract

Four general definitions of Machine Learning (ML), from well-know machine learners, are first provided as an approximation to define its main objective. Siimilarly, then three general definitions of Pattern Recognition (PR), by three renowned authors, are discussed. These discussions are followed be an introduction of the main paradigm for ML/PR systems: the classification paradigm. It is illustrated by a simple OCR example, which is also the used to describe the conventional structure of a classifier and the two conventional learning methods. Later, a few, outstanding application examples of ML/PR systems are given, and the presentation ends by providing references to cited authors.

The training objectives are: 1) To define machine learning and pattern recognition; 2) To interpret the classification paradigm and the conventional classifier structure; 3) To understand the conventional learning methods; and 4) To know some pattern recognition applications.