A GA-Based Fuzzy Decision Tree Approach for Corporate Bond Rating

摘要: The induction based on a tree structure is an appropriate representation of the complex human reasoning process such as corporate bond rating application. Furthermore, fuzzy decision (FDT) can handle information about vague and incomplete classification knowledge represented in linguistic terms. In addition, FDT more flexible by relaxing constraint mutual exclusivity cases tree. We propose hybrid approach using genetic algorithms (GA) enhances effectiveness to problem classification. This study utilizes GA attempt find optimal or near hurdle values membership function FDT. results show that accuracy integrated proposed for this increases overall rate significantly. also flexibility process.