Existing literature indicates that social backgrounds influence occupational choices, making them a crucial consideration for economic and human capital policies. The gig work has emerged as a popular alternative form of labor, with individuals leveraging technologies to seek and perform tasks. Gig workers play a vital role in advancing socio-economic agendas in many countries, prompting the introduction of various policies to nurture the ecosystem. Given its novelty, research on the characteristics of gig workers remains limited. With big data and machine learning analytics, it is now possible to integrate traditional approaches with new techniques in research. In this study, we aim to investigate the gig workforce by pursuing two objectives: to identify the profiles and job-seeking patterns of gig workers and to determine their work and earning behaviors within a selected segment of the gig economy. We employ a combination of data analytics and empirical surveys, focusing on data from Malaysia. Specifically, this study examined over 400,000 user profiles on the nation’s gig economy platform eRezeki via unsupervised machine learning (clustering) and association rule mining to unveil behavioral and demographic patterns. K-means clustering divided gig workers into groups according to their demographic and skill traits, and association rule mining indicated frequent job preference and income patterns related to user attributes. A further 300 Kuala Lumpur, Selangor, and Putrajaya delivery riders were also given guided questionnaires to complement and validate the platform data. This study contributes in two ways. It enhances the understanding of the talent profiles of gig workers in relation to national strategies for the gig economy. Furthermore, it illustrates the potential and complementary roles of data mining and empirical surveys in conducting research. We discuss several implications for future research.
Fundamental Research Grant (FRGS) Malaysia and Malaysia Economic Development Corporation (MDEC).
| Primary Language | English |
|---|---|
| Subjects | Labor Sociology |
| Journal Section | Research Article |
| Authors | |
| Submission Date | October 16, 2024 |
| Acceptance Date | August 22, 2025 |
| Publication Date | December 15, 2025 |
| DOI | https://doi.org/10.26650/JECS2024-1565181 |
| IZ | https://izlik.org/JA75RA75WL |
| Published in Issue | Year 2025 Issue: 72 |