Improvised Harris Hawks Optimization With Rough Set Min-Max Classifier for Feature Selection to Enhance Disease Prediction

Devi S; V.S. Arulmurugan; V Sujatha; K.E. Narayana; Saranya K S; M. Balamurugan; J Parthasarathy

doi:10.5935/jetia.v12i59.3402

Authors

Devi S Department of Master of Computer Applications, Nandha Engineeering College, Erode, India. https://orcid.org/0000-0002-7472-2319 (unauthenticated)
V.S. Arulmurugan Professor & Head, Department of Electrical and Electronics Engineering, Shree Venkateshwara Hi-Tech Engineering College (Autonomous), Gobichettipalayam, Erode, India. https://orcid.org/0000-0002-2571-454X (unauthenticated)
V Sujatha Professor, Department of Electronics and Communication Engineering, Shree Sathyam College of Engineering and Technology, Sankari, India.
K.E. Narayana Assistant Professor, Department of Computer Science and Engineering, Rajalakshmi Engineering College, Chennai, India. https://orcid.org/0000-0002-4132-2175 (unauthenticated)
Saranya K S Associate Professor, Department of Computer Science and Engineering, Bannari Amman Institute of Technology, Sathya Mangalam, India. https://orcid.org/0009-0003-3217-5817 (unauthenticated)
M. Balamurugan Head of the Department & Associate Professor, Department of Computer Science and Engineering, The Kavery Engineering College, Salem, India https://orcid.org/0009-0004-0007-9406 (unauthenticated)
J Parthasarathy Professor, Department of Information Technology, AVS Engineering College, Salem, India. https://orcid.org/0009-0006-7634-7045 (unauthenticated)

DOI:

https://doi.org/10.5935/jetia.v12i59.3402

Abstract

The significant development in the medical field emerges a massive production of medical data, which must be appropriately analyzed to determine any disease. This sensitive information contains lots of redundancy or noise data which needs to be reduced through the Feature Selection (FS) approach. The researchers need an optimal system to solve this dimensionality reduction problem. Therefore, we proposed a novel Improvised Harris Hawks Optimization (IHHO) algorithm based on Rough Set Min-Max Classifier (RMMC) to solve the FS problem using Chaotic-Improved Lévy Flight Distribution. IHHO is used to search for the best feature subset, whereas RMMC is used to design a clever rule for searching features using Euclidean distance measurement. Nearly three different datasets (Epileptic Seizure Recognition Data Set, Liver Dataset, and Breast Cancer dataset) are utilized to evaluate the performance of the proposed IHHO-RMMC model. According to our experimental results, the proposed IHHO-RMMC approach can solve FS problems and achieves 99.65% accuracy for Epileptic Seizure Recognition Data Set, 98.12% accuracy for Breast Cancer Dataset, and 96.22% accuracy for Liver Dataset which is higher than existing RMMC methods. The proposed IHHO-RMMC is expected to offer a high degree of accuracy in predicting diseases in the medical field.