ISBN-13: 9783639710991 / Angielski / Miękka / 2014 / 248 str.
The book begins with discussion on the basic concept on Data Mining, emphasizing the need of classification model. However, it mainly focuses on designing a family of new hybrid classification systems, each combining C4.5 (a decision tree based rule inductive algorithm) and genetic algorithm. Formally, each such system consists of three phases. The first phase attempts to produce a good population (rule set) from training set. The second phase resolves the interpretability problem of the population and, finally GA optimizes the formatted rule set. The ultimate aim of each system is to achieve higher prediction accuracy over classification problem irrespective to domain, size, dimensionality and class distribution, accepting a good population learned by C4.5 at the beginning. Certainly, the book is not only useful to the researchers but also helpful to the undergraduate and postgraduate students of computer science.
The book begins with discussion on the basic concept on Data Mining, emphasizing the need of classification model. However, it mainly focuses on designing a family of new hybrid classification systems, each combining C4.5 (a decision tree based rule inductive algorithm) and genetic algorithm. Formally, each such system consists of three phases. The first phase attempts to produce a good population (rule set) from training set. The second phase resolves the interpretability problem of the population and, finally GA optimizes the formatted rule set. The ultimate aim of each system is to achieve higher prediction accuracy over classification problem irrespective to domain, size, dimensionality and class distribution, accepting a good population learned by C4.5 at the beginning. Certainly, the book is not only useful to the researchers but also helpful to the undergraduate and postgraduate students of computer science.