JobKorea, an AI and data-based HR tech platform operated by Worksphere, said on Thursday it has confirmed changes in jobseeker behaviour and conversion performance in a short period after applying an upgraded version of its next-generation personalised recommendation system, AI Recommendation 3.0.

AI Recommendation 3.0 is the first case of implementing in an actual service the “Context Link” presented as a key agenda at a recent 30th anniversary event. It reflects jobseeker interests, behaviour and browsing context in real time to suggest postings needed at the moment.

The company said an analysis of operating data over about 10 days after the upgrade showed big improvements in jobseekers’ efficiency in exploring postings and in application conversion. Click-through rates on saved postings and the share of job applications rose about 300 percent versus the average of the previous recommendation version, it said. The company said it confirmed the effect of recommendations that lead to real actions rather than simple exposure.

The company stressed the results are significant because they stem from a shift in recommendation structure aligned with changing jobseeker needs, rather than a simple addition of functions.

AI Recommendation 3.0 is designed to comprehensively learn each user’s behavioural context, including which postings a jobseeker checked, where they stayed and which jobs and keywords they repeatedly responded to. It is built to connect postings that match the browsing flow.

AI Recommendation 3.0 is applied to the “Today’s AI Insight” and “Curation JOB” areas on JobKorea’s main screen and is provided to all users. It combines resume information with actual browsing and application behaviour to selectively suggest only the information jobseekers need. The company said it believes conversion performance was also affected by UI and UX improvements, including the introduction of left-right scrolling, which reduced unnecessary searches and strengthened a recommendation-centred browsing experience.

JobKorea plans to accelerate upgrades to a Context Link-based recommendation model that connects personal behavioural data to corporate reviews, organisational culture and work-experience context. Its strategy is to expand into a recommendation service that goes beyond simple condition matching and connects “experience-unit choices” that jobseekers are likely to prefer. It also plans to apply verified AI recommendation insights to Albamon in phases to strengthen a context-based recommendation model suited to the characteristics of the part-time job market.

Yoseop Kim (김요섭), JobKorea’s chief technology officer, said, “Going beyond simply showing many postings, AI Recommendation 3.0 is a recommendation model that understands a jobseeker’s ‘current situation’ and ‘context of interest’ and helps choices that lead to real actions.” He added, “As it has been validated with results in the short period of 10 days, we will continue to raise the standard of the job-search experience by advancing Context Link-based recommendations.”