1. K. S. Rangasamy College of Technology, Tiruchengode, Tamil Nādu, India
2. FPT Greenwich University, Ho Chi Minh City, Vietnam
Email: sarumathi@ksrct.ac.in (S.S.); reshmafarhana17@gmail.com (N.R.); SharmilaM@fe.edu.vn (Sh.M.); malarkhodi@ksrct.ac.in (S.M.)
*Corresponding author
Manuscript received June 25, 2023; revised September 15, 2023; accepted May 14, 2024; published August 28, 2024
Abstract—Alzheimer’s Disease (AD) is one of the prevalent diseases which is a neurological condition that impairs brain activities like reading, writing, thinking, and remembering. The death rate due to AD would be reduced by providing proper treatment based on the stage of the disease. This can be determined by using data mining techniques. A data mining technique, Binary version of the Artificial Bee Colony (BABC) algorithm was proposed to choose the best features from statistical and volumetric information of Magnetic Resonance Images (MRIs) of the brain. However, the accuracy of BABC is low due to slow convergence. So, in this article, an Improved Artificial Bee Colony (IABC) algorithm is introduced to enhance the AD prediction accuracy. It can be achieved by improving the exploration and exploitation process of BABC. In the employee bee phase of IABC, a novel search equation is used that enhances the probabilities for onlookers’ bees to determine the best positions and change the number of bad ones by the fresh ones in the following phase. Furthermore, Particle Swarm Optimization (PSO) is utilized to create a fresh position changing an un-updated location in the scout bee phase. IABC is enhancing the AD prediction efficiency and interpretability by identifying the most relevant predictors, reducing dimensionality, and improving model generalization for the AD prediction. Furthermore, it also improves the exploration-exploitation process of feature selection. From the empirical findings, it is proved that the proposed IABC with Random Forest (IABC-RF) has 10.52%, 8.57%, 7.87%, and 6.8% better accuracy, precision, recall and F-measure than BABC with K-Nearest Neighbor (BABC-KNN) for the AD prediction.
Keywords—Alzheimer Disease (AD), Artificial Bee Colony (ABC) algorithm, feature selection, Improved Artificial Bee Colony (IABC) algorithm
[PDF]
Cite: S. Sarumathi, N. Reshma, Sharmila Mathivanan, and S. Malarkhodi, " An Improved Attribute Subset Selector for Alzheimer’s Disease Prediction," International Journal of Computer Theory and Engineering, vol. 16, no. 3, pp. 87-93, 2024.
Copyright © 2024 by the authors. 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 (CC BY 4.0).
