Chapter 5 – Machine learning

5.1  Overview

5.2  Why Do We Want Machine Learning?

5.3  How Machines Learn

5.3.1  Phases of Machine Learning

5.3.2  Rote Learning and the Importance of Generalisation

5.3.3  Inputs to Training

5.3.4  Outputs of Training

5.3.5  The Training Process

5.4  Deductive Learning

5.5  Inductive Learning

5.5.1  Version Spaces

5.5.2  Decision Trees

5.5.2.1  Building a Binary Tree

5.5.2.2  More Complex Trees

5.5.3  Rule Induction and Credit Assignment

5.6  Explanation-Based Learning

5.7  Example: Query-by-Browsing

5.7.1  What the User Sees

5.7.2  How It Works

5.7.3  Problems

5.8  Summary

abductive reasoning, application phase, bias, branch and bound search, C4.5, case-based reasoning, clustering, computer chess, concept learning, contingency table, credit assignment, database, decision table, decision tree, deductive learning, deductive reasoning, domain-independent knowledge, entropy, expert system, explanation-based learning, genetic algorithm, ID3, inductive learning, inductive reasoning, knowledge elicitation, Lex, memorising, Occam's razor, overfitting, pole balancing, predicate logic, pruning, Query-by-Browsing, random forest, reasoning by analogy, regret, relational database, robotics, rote learning, rule induction, salience, search, self-organising map, semantic network, SOAR, supervised learning, training phase, unsupervised learning, validation phase, version-space algorithm