Diabetes Classification using an Expert Neuro-fuzzy Feature Extraction Model

Diabetes is one of the challenging diseases prevailing in recent times. Due to the incompleteness, uncertainty and imprecise details, classification of diabetes using machine learning algorithms is turning out to be even more challenging. The efficiency of the classification model is influenced by the data present in the dataset. This study enhances the classification of diabetes by using a Neuro-Fuzzy model with special attention to Feature Extraction. The main goal of the present study is to enhance the diabetes prediction technique that helps the medical practitioners to easily identify the disease and diagnose it appropriately to reduce several complications that diabetes may cause to the patient in the future. The proposed model initially applies fuzzification on diabetes data to produce membership values. Later the membership values are examined by the proposed model to check the contribution of the features in diabetes classification. The feature extraction algorithm passes the significant features to a neural network after the features are extracted. The proposed model is tested on standard PIMA diabetic dataset to evaluate the performance. The proposed model is able to outperform all the existing machine learning algorithms.

@article{Chowdary2021,
  title     = {Diabetes Classification using an Expert Neuro-fuzzy Feature Extraction Model},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {12},
  number    = {8},
  year      = {2021},
  publisher = {The Science and Information Organization},
  author    = {P. Bharath Kumar Chowdary and R. Udaya Kumar},
  doi       = {10.14569/IJACSA.2021.0120842},
  url       = {https://doi.org/10.14569/IJACSA.2021.0120842}
}