Da-in Choi, Moohong Min, Journal of Internet Computing and Services, Vol. 26, No. 3, pp. 99-109, Jun. 2025
10.7472/jksii.2025.26.3.99, Full Text:  HTML
Keywords: cybersecurity, Cybersecurity Workforce, Job Analysis, Unsupervised learning, topic modeling

Abstract

This study proposed a method to provide more timely statistics on cybersecurity workforce trends by utilizing online job postings. Traditional cyber security workforce surveys, such as the Information Security Industry Survey and the Information Security Survey, have limitations in capturing workforce distribution and detailed job classifications within companies. These surveys are conducted annually, resulting in long delays from data collection to publication, reducing their timeliness. To address these limitations, this study introduces a methodology that leverages job postings from recruitment platforms. Each job posting is segmented into specific job roles for a more granular analysis. Unsupervised learning techniques are applied to cluster the data, which is then matched to the National Competency Standards(NCS) job classification framework to generate workforce statistics. This approach enables the real-time monitoring of cybersecurity workforce trends and provides more timely, data-driven insights. Ultimately, it can assist in formulating and implementing workforce development policies aligned with industry demands.

Statistics
Cite this article