Optimized Fuzzy Decision Tree for Structured Continuous-Label Classification

Mainly understandable decision trees have been intended for perfect symbolic data. Conventional crisp decision trees (DT) are extensively used for classification purpose. However, there are still many issues particularly when we used the numerical (continuous valued) attributes. Structured continuouslabel classification is one type of classification in which the label is continuous in the data. Although, the main aim is to classify data into classes that are a set of predefined ranges and can be ordered in a hierarchy. Some of those difficulties can be addressed using fuzzy decision trees (FDT). Over the years, additional methodologies have been investigated and proposed to deal with continuous or multi-valued data, and with missing or noisy features. In recent times, with the growing popularity of fuzzy representation, a few researchers independently have proposed to utilize fuzzy representation in decision trees to deal with similar situations. Fuzzy concept connects the gap between symbolic and non symbolic data by using linking qualitative linguistic terms with quantitative data. In our proposed work, a new method of optimized fuzzy decision trees is presented. This method proposed a novel decision tree technique for handling continuous valued attributes with user defined membership. The efficient results of our proposed fuzzy decision trees are compared at the end in the experimentation. From the experimentation, we are conclude that the proposed fuzzy decision trees is well effective than the existing system interms of accuracy rate as well as reduce the error performance of the system.