This chapter aims to give a comprehensive view about the links between fuzzy logic and data mining. It will be shown that knowledge extracted from simple data sets or huge databases can be represented by fuzzy rule-based expert systems. It is highlighted that both model performance and interpretability of the mined fuzzy models are of major importance, and effort is required to keep the resulting rule bases small and comprehensible. Therefore, in the previous years, soft computing based data mining algorithms have been developed for feature selection, feature extraction, model optimization, and model reduction (rule based simplification). Application of these techniques is illustrated using the wine data classification problem. The results illustrate that fuzzy tools can be applied in a synergistic manner through the nine steps of knowledge discovery.
Key Terms in this Chapter
Fuzzy Clustering: A family of methods that aims to partition the data set into clusters in a way that objects are allowed to belong to several clusters simultaneously with different degrees of membership.
Rule Base Reduction: Aims to discover the redundant or unimportant rules to simplify the rule based model.
Fuzzy Classifier System: A fuzzy rule based model that can be used to obtain whether a given pattern in which class should be classified.
Knowledge Discovery in Databases: Refers to the overall process of discovering knowledge from data.
Visualization: A technique that can be used to map high-dimensional data or other objects like clusters into two- or three-dimensional space.
Association Rule Mining: Aims to discover dependencies between attributes on the basis of frequent item sets extracted from the measured data.