An Efficient Genetic Disorder Classification Framework Using Feature Selection, Machine Learning and Deep Learning Techniques
Year: 2026
Type: conference-paper
Abstract: Accurate prediction of genetic disorders is a critical task in medical diagnosis due to the complexity and high dimensionality of genomic and clinical data. Traditional approaches often struggle to handle irrelevant and redundant features, leading to reduced classification performance. This study presents an optimized framework for genetic disorder prediction using feature selection and machine learning techniques. Initially, multiple machine learning models are applied to the complete dataset to establish baseline performance. Subsequently, feature selection methods such as Chi-Square and Mutual Information are employed to identify the most relevant attributes. The selected features are then used to train both ML and DL models, including an Artificial Neural Network. The experimental results demonstrate that feature selection significantly enhances classification performance and reduces computational complexity. Among the evaluated approaches, the deep learning model achieved superior performance, indicating its effectiveness in capturing complex patterns in the dataset. The proposed framework provides a reliable and scalable solution for improving genetic disorder prediction.
Authors: Ravada Sony, Chiranjeevi Manike
Institution: SR University
Language: English
Subfield: Health Information Management
Field: Health Professions
Domain: Health Sciences
Open Access status: hybrid
OpenAlex ID: W7204149551