Work

Machine Learning for Community Well-Being: Identifying Factors Affecting Resilience in Disaster-Prone Regions

Year: 2026

Type: article

Abstract: This paper presents a machine learning approach to identify factors affecting community well-being in New South Wales (NSW), Australia, based on regional well-being survey (RWS) data. The study focuses on NSW, which has recently experienced multiple types of disasters, and utilizes machine learning techniques to identify indicators that can help prevent the decline of community well-being using community users’ responses to survey data. However, obtaining labelled data for this type of study is challenging, so the paper introduces a recommender-based approach that utilizes randomly generated instances, eliminating the need for labels to identify factors. The primary objective is to investigate the factors that affect the current level of community well-being. The study identifies these factors using a recommender-based model and highlights the adaptability of this model across regional well-being survey datasets, indicating its potential to make significant contributions to the field of disaster resilience and enhance community well-being.

Source: International Journal of Community Well-Being

DOI: 10.1007/s42413-026-00305-3

Authors: Nasrin Sultana, Md Sarwar Kamal, Ho Leung IP, Michael Bewong, Md Zahidul Islam, Azizur Rahman

Institutions: RMIT University, Charles Sturt University, Auckland University of Technology

Language: English

Topic: Disaster Management and Resilience

Subfield: Sociology and Political Science

Field: Social Sciences

Domain: Social Sciences

SDGs: Good health and well-being, Sustainable cities and communities

Open Access status: hybrid

Estimated APC paid: $3,290