Data Mining: A Heuristic Approach

Real life problems are known to be messy, dynamic and multi-objective, and involve high levels of uncertainty and constraints. Because traditional problem-solving methods are no longer capable of handling this level of complexity, heuristic search methods have attracted increasing attention in recent years for solving such problems. Inspired by nature, biology, statistical mechanics, physics and neuroscience, heuristics techniques are used to solve many problems where traditional methods have failed. Data Mining: A Heuristic Approach will be a repository for the applications of these techniques in the area of data mining.

Real-life problems are known to be messy, dynamic and multi-objective, and involve high levels of uncertainty and constraints. Because traditional problem-solving methods are no longer capable of handling this level of complexity, heuristic search methods have attracted increasing attention in recent years for solving such problems. Inspired by nature, biology, statistical mechanics, physics and neuroscience, heuristic techniques are used to solve many problems where traditional methods have failed. Data Mining: A Heuristic Approach is a repository for the applications of these techniques in the area of data mining.

Booknews

This collection applies heuristic search methods to the data mining of computer databases. The 13 papers illustrate the potential role of evolutionary algorithms, genetic programming, ant colony optimization, and artificial immune systems. Some of the topics are the estimation of distribution algorithms for feature subset selection on large dimensionality domains, use of the hill climbing method for clustering, an ant colony algorithm for classification rule discovery, and parallel data mining. An opening chapter provides the mathematical background for five heuristic approaches. Annotation c. Book News, Inc., Portland, OR (booknews.com)