Fuzzy Modeling and Genetic Algorithms for Data Mining and Exploration

3.7 Neural Networks

■

51

TABLE 3.6

Strengths/Weaknesses of Decision Trees and Classifiers

Strengths

■

Creates a set of classification rules

■

Good for multidimensional data

■

Easy-to-visualize dependency relationships

■

Several tools available for decision trees (and case-based reasoning)

■

Low computational requirements

■

Organization sensitive to information gain (i.e., the amount of knowl-

edge exposed by the tree)

■

Usually capable of handling both continuous and categorical data

■

Probabilities on Markov chain can highlight important relationships and

depress rules created by spurious outliers

■

Nearly always identify which variables make a significant contribution ...