Abstract
Due to the fact that healthcare applications produce a lot of data, it is different in terms of volume, diversity, velocity, veracity, and value. There will be issues with the classification of healthcare data as a result of the massive collection of medical data. Therefore, effective big data processing techniques are quickly needed for this type of data. Many of these issues have been effectively resolved using fuzzy systems. This method offers a new way for handling big data that uses a Fuzzy rule-based Convolutional Neural Network (FCNN) classifier to categorize the big data generated in this environment for healthcare services. In this analysis, a Fuzzy Convolutional Neural Network (FCNN) model is used to categorize healthcare data. The information from the big data is gathered by this model, which also does preprocessing, clustering, feature selection, and classification of the data. Using the principal component analysis algorithm, the attributes that are unrelated are eliminated. The presented technique utilizes a clustering strategy based on fuzzy rules. The classification of normal and disease-related data is then efficiently decided using an FCNN classifier. This approach is evaluated using a number of evaluation criteria, including precision, recall, accuracy, and F1-Score. The outcomes gained support the presented scheme's efficacy in relation to several performance evaluation parameters.
Keyword
Healthcare, Big Data, Fuzzy Rule based classifier, Expectation-Maximization (EM).
PDF Download (click here)
|