Article: Autism spectrum disorder prediction using LASSO regularised bat search optimisation Journal: International Journal of Services Operations and Informatics (IJSOI) 2024 Vol.13 No.1 pp.58 - 74 Abstract: Autism spectrum disorder (ASD) is a neuro developmental disorder characterised by persistent social interaction, communication impairments, restricted and repetitive behaviour patterns. Early and accurate diagnosis of ASD is crucial for effective intervention and support. Over the past decade, machine learning (ML) techniques have shown promise in aiding ASD diagnosis by leveraging large-scale datasets and identifying clinical data patterns. This paper proposes an improved swarm algorithm, a least absolute shrinkage and selection operator (LASSO) regularised bat search optimisation algorithm (LBSO), to improve the predictive performance of ASD diagnosis. The LASSO-based bat search optimisation (LBSO) algorithm integrates the LASSO regularisation technique into the BSO algorithm to enhance feature selection and optimisation. LASSO effectively reduces the impact of irrelevant or redundant features, promoting sparsity and improving model interpretability. The BSO algorithm, inspired by the echolocation behaviour of bats, offers efficient search and optimisation capabilities. Inderscience Publishers - linking academia, business and industry through research

Title: Autism spectrum disorder prediction using LASSO regularised bat search optimisation

Authors: Keerthi Guttikonda; G. Ramachandran; G.V.S.N.R.V. Prasad

Addresses: Department of Computer Science and Engineering, Annamalai University, Annamalainagar, 608002, Tamilnadu, India ' Department of Computer Science and Engineering, Annamalai University, Annamalainagar, 608002, Tamilnadu, India ' Department of Computer Science and Engineering, Gudlavalleru Engineering College, Gudlavalleru – 521356, Andhra Pradesh, India

Abstract: Autism spectrum disorder (ASD) is a neuro developmental disorder characterised by persistent social interaction, communication impairments, restricted and repetitive behaviour patterns. Early and accurate diagnosis of ASD is crucial for effective intervention and support. Over the past decade, machine learning (ML) techniques have shown promise in aiding ASD diagnosis by leveraging large-scale datasets and identifying clinical data patterns. This paper proposes an improved swarm algorithm, a least absolute shrinkage and selection operator (LASSO) regularised bat search optimisation algorithm (LBSO), to improve the predictive performance of ASD diagnosis. The LASSO-based bat search optimisation (LBSO) algorithm integrates the LASSO regularisation technique into the BSO algorithm to enhance feature selection and optimisation. LASSO effectively reduces the impact of irrelevant or redundant features, promoting sparsity and improving model interpretability. The BSO algorithm, inspired by the echolocation behaviour of bats, offers efficient search and optimisation capabilities.

Keywords: ASD; autism spectrum disorder; machine learning; diagnosis; feature selection; bat algorithm; regularisation.

DOI: 10.1504/IJSOI.2024.143149

International Journal of Services Operations and Informatics, 2024 Vol.13 No.1, pp.58 - 74

Received: 21 Dec 2023
Accepted: 31 Dec 2023
Published online: 04 Dec 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Autism spectrum disorder prediction using LASSO regularised bat search optimisation

Keep up-to-date