Advancing Autism Spectrum Disorder Diagnosis through Ensemble Learning

Nishat Tanveen*, Juvvaladinne Trisha**, Ambatipudi Durga Sravani***
*-*** Department of Computer Science and Engineering, Vasireddy Venkatadri Institute of Technology, Nambur, Guntur, Andhra Pradesh, India.
Periodicity:January - June'2024
DOI : https://doi.org/10.26634/jaim.2.1.20500

Abstract

Autism Spectrum Disorder (ASD) diagnostics requires specialized clinical expertise, posing accessibility and affordability barriers for many. To widen the availability of precise screening, this paper examines ensemble machine learning models that combine multiple algorithms for improved accuracy and generalizability. Specifically, this paper compares the performance of K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Naive Bayes, and a Voting Classifier that integrates Logistic Regression, Decision Trees, and KNN. The comparison was conducted on separate datasets of pediatric and adult ASD questionnaire responses. The ensemble model significantly outperforms individual techniques, achieving higher accuracy for both pediatrics and adults with balanced sensitivity and specificity maintained across groups, indicating the viability of accessible community-available screening to alleviate diagnostic bottlenecks. Before scale-up, further model optimization for interpretability and testing on more diverse multi-site data are warranted. Overall, findings demonstrate the feasibility of mobile distributed pre-screening systems leveraging optimized ensembles to predict ASD with high precision across ages, opening possibilities for explainable AI to lower costs and widen access compared to in-person evaluation.

Keywords

Autism Spectrum Disorder, Ensemble Learning, Multiple Algorithms, K-Nearest Neighbor, Support Vector Machine, Naive Bayes, Voting Classifier.

How to Cite this Article?

Tanveen, N., Trisha, J., and Sravani, A. D. (2024). Advancing Autism Spectrum Disorder Diagnosis through Ensemble Learning. i-manager’s Journal on Artificial Intelligence & Machine Learning, 2(1), 28-36. https://doi.org/10.26634/jaim.2.1.20500

References

[7]. Schlink, A. J. (2022). Towards the Improved Characterization of Minimally Verbal Children with Autism: Applications of Item Response Theory and Machine Learning Algorithms to Analyze Measures of Social Communication. University of California, Los Angeles.
If you have access to this article please login to view the article or kindly login to purchase the article

Purchase Instant Access

Single Article

North Americas,UK,
Middle East,Europe
India Rest of world
USD EUR INR USD-ROW
Pdf 40 40 300
Online 40 40 300
Pdf & Online 40 40 300

Options for accessing this content:
  • If you would like institutional access to this content, please recommend the title to your librarian.
    Library Recommendation Form
  • If you already have i-manager's user account: Login above and proceed to purchase the article.
  • New Users: Please register, then proceed to purchase the article.