Chellappan, D.; Rajaguru, H. Enhancement of Classifier Performance Using Swarm Intelligence in Detection of Diabetes from Pancreatic Microarray Gene Data. Biomimetics2023, 8, 503.
Chellappan, D.; Rajaguru, H. Enhancement of Classifier Performance Using Swarm Intelligence in Detection of Diabetes from Pancreatic Microarray Gene Data. Biomimetics 2023, 8, 503.
Chellappan, D.; Rajaguru, H. Enhancement of Classifier Performance Using Swarm Intelligence in Detection of Diabetes from Pancreatic Microarray Gene Data. Biomimetics2023, 8, 503.
Chellappan, D.; Rajaguru, H. Enhancement of Classifier Performance Using Swarm Intelligence in Detection of Diabetes from Pancreatic Microarray Gene Data. Biomimetics 2023, 8, 503.
Abstract
Diabetes mellitus is a chronic disease that affects millions of people worldwide. Article focuses on detecting the presence of diabetes in patients using microarray gene data obtained from the pancreas. Handle the high-dimensional nature of the data, four different dimensionality reduction techniques, namely Bessel Function, Discrete Cosine Transform (DCT), Least Square Linear Regression (LSLR), and Artificial Algae Algorithm (AAA) are used. After reduced the data, Meta-heuristic algorithms are applied like Dragonfly Optimization Algorithm (DOA) and Elephant Herding Optimization Algorithm (EHO) for feature selection. Ten classification techniques are using to classify the data in both the format like without feature selection method and with feature selection method. The classification techniques are Non-Linear Regression (NLR), Linear Regression (LR), Gaussian Mixture Model (GMM), Expectation Maximum (EM), Bayesian Linear Discriminant Classifier (BLDC), Logistic Regression (LoR), Softmax Discriminant Classifier (SDC), Support Vector Machine (SVM) with linear kernel, Support Vector Machine (SVM) with polynomial kernel, and Support Vector Machine (SVM) with Radial Basis Function (RBF) kernel. Results showed that the AAA with SVM(RBF) achieved an accuracy of 90% without feature selection. However, when feature selection was applied, with EHO of AAA with SVM(RBF) exhibited the highest accuracy of 95.714%, followed closely by with DOA of AAA with SVM (RBF) at 94.28%.
Keywords
Type II Diabetes Mellitus (DM); microarray gene data; Dimensionality Reduction (DR); Classification techniques; feature selection; LR; AAA; DOA; EHO
Subject
Engineering, Bioengineering
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
