Framework for the development of data-driven mamdani-type fuzzy decision support systems based on fuzzy set theory using clusters and pivot tables

The objectives of this thesis were to design, implement and validate a framework for the development of data-driven Mamdani-type fuzzy Decision support systems using clusters and pivot tables. For the first objective we designed an architecture of layers with their components for their respective implementation. The second objective was to develop an efficient implementation of each of the designed layers with their respective components in the MATLAB platform. The third objective was to provide an efficient validation of the complete framework in real-world classification problems for validating if it could extract significant features and the knowledge base and rule base for the development of Data-driven Mamdani-type Decision support systems. For validating the proposed methodology, we applied our algorithms on four public datasets including Wisconsin, and Coimbra breast cancer, wart treatment, and compared them with other related works obtained from the literature. According to the obtained results it could be demonstrated that the knowledge database and the knowledge rule base could be extracted transparently. The bene its of these expert systems are multifold: assisting physicians in selecting the best treatment method, saving time for patients, reducing the treatment cost, and improving the quality of treatment.