Diabetes disables body to regulate proper amount of glucose as insulin. It has impacted a vast global population. In this paper, we demonstrated a fuzzy c-means-neuro-fuzzy rule-based classifier to detect diabetic disease with an acceptable interpretability. We measured the accuracy of our implemented classifier by correctly recognizing diabetic records. Besides we measured the complexity of the classifiers by the number of selected fuzzy rules. To achieve good accuracy and interpretability, the implemented fuzzy classifier can be treated as an acceptable trade-off. At the end of the research, we compared our experiment results with the achieved results from certain medical institutions that worked on the same type of dataset which demonstrated the compactness, accuracy of the proposed approach.
CITATION STYLE
Kabir, H., Ridwan, S. N., Mosharof Hossain, A. T. M., Tuktuki, N. H., Haque, F., Afrin, F., & Rahman, R. M. (2016). Application of fuzzy logic for generating interpretable pattern for diabetes disease in Bangladesh. In Advances in Intelligent Systems and Computing (Vol. 464, pp. 403–413). Springer Verlag. https://doi.org/10.1007/978-3-319-33625-1_36
Mendeley helps you to discover research relevant for your work.