No abstract

No abstract

Type-Diabetes Mellitus (T2DM), a major threat to developing as well as developed countries, can be easily controlled to a large extent through lifestyle modifications. Diabetes increases the risk of developing various health as well as financial problems to cure these health complications. The health complications are stroke, myocardial infarction, and coronary artery disease. Nerve, muscle, kidney and retinal damage have distressing impact on the life of a diabetic patient. It is the need of the hour to halt the epidemic of T2DM in the early stage. Data science approaches have the potential to predict on medical data. Machine learning is an evolving scientific field in data science where machines learn mechanically and improve from experience without any explicit program. Our goal was to develop a system which can improve performance of a classifier for prediction of T2DM. The purpose of this work is to implement a hybrid model for prediction by integrating the advantages of artificial neural net (ANN) and fuzzy logic. Genetic algorithm (GA) and particle swarm optimization (PSO) have been applied to optimize parameters of developed predicting model. The proposed scheme used a fuzzification matrix. This matrix is used to relate the input patterns with a degree of membership to different classes. The specific class is predicted based on the value of degree of membership of a pattern. We have analyzed the proposed method and previous research in the literature. High accuracy was achieved using the ANFIS-PSO approach.

No abstract

In today's digital world, a dataset with large number of attributes has a curse of dimensionality where the computation time grows exponentially with the number of dimensions. To overcome the problem of computation time and space, appropriate method of feature selection can be developed using metaheuristic approaches. The aim of this work is to investigate the use of ant colony optimization with the help of neural network to select near optimal feature subset and integrate it with the self-organizing fuzzy logic classifier for improving the recognition rate. The proposed fuzzy classifier derives prototype from the collected data through an offline training process and uses it to develop a fuzzy inference system for classification. Once trained, it can continuously learn from streaming data and later adapts the changing facts by updating the system structure recursively. The developed model is not based on predefined parameters used in the data generation model but is derived from the empirically observed data.