MACHINE LEARNING TECHNIQUES FOR IDENTIFYING SELF-CARE PROBLEMS IN DISABLED CHILDREN

Abstract

The identification and intervention of self-care problems in disabled children are crucial for enhancing their quality of life and independence. The utilization of machine learning algorithms holds promise in revolutionizing the identification and handling of these self-care challenges potentially offering tailored solutions to improve the well-being and autonomy of disabled children. Therefore, a literature review is imperative to comprehensively assess the landscape of machine learning (ML) applications in addressing these self-care challenges. Existing SLRs on this topic lack comprehensive coverage of ML-based techniques, hindering a full understanding of their efficacy in classifying self-care issues among disabled children. This review aims to assess that how ML methodologies contribute to identify and address these challenges along with their impacts on accuracy and clinical relevance. By encapsulating various ML methodologies used in diagnosing selfcare problems, this review reveals their diverse impacts on accuracy and clinical applicability. The novel aspect of this work lies in the comprehensive coverage and evaluation of diverse ML techniques, highlighting their potential to transform pediatric healthcare for disabled children. In conclusion, this review demonstrates that hybrid ML models, feature selection and extraction techniques significantly enhance classification accuracy paving the way for improved interventions. This comprehensive analysis makes this review a valuable resource for researchers seeking insights into ML's role in addressing self-care challenges among disabled children.