Course Recommender Systems Need to Consider the Job Market

TL;DR

This work tackles the mismatch between evolving job-market skill demands and traditional course recommender systems. It proposes a job-market-oriented framework that combines unsupervised skill extraction via LLM prompts (SEM) with RL-driven sequential course recommendations (JCRec), grounded in the ESCO skill taxonomy. The paper formalizes a set of properties (P1–P5) and research directions (RD1–RD6) to advance the field and presents empirical evidence using open datasets, showing that PPO can efficiently deliver high-quality long-sequence recommendations. The proposed approach has potential economic and societal impact by guiding learners toward in-demand skills and roles while remaining adaptable to market changes.

Abstract

Current course recommender systems primarily leverage learner-course interactions, course content, learner preferences, and supplementary course details like instructor, institution, ratings, and reviews, to make their recommendation. However, these systems often overlook a critical aspect: the evolving skill demand of the job market. This paper focuses on the perspective of academic researchers, working in collaboration with the industry, aiming to develop a course recommender system that incorporates job market skill demands. In light of the job market's rapid changes and the current state of research in course recommender systems, we outline essential properties for course recommender systems to address these demands effectively, including explainable, sequential, unsupervised, and aligned with the job market and user's goals. Our discussion extends to the challenges and research questions this objective entails, including unsupervised skill extraction from job listings, course descriptions, and resumes, as well as predicting recommendations that align with learner objectives and the job market and designing metrics to evaluate this alignment. Furthermore, we introduce an initial system that addresses some existing limitations of course recommender systems using large Language Models (LLMs) for skill extraction and Reinforcement Learning (RL) for alignment with the job market. We provide empirical results using open-source data to demonstrate its effectiveness.

Figures (2)

• Figure 1: Illustration of the skill extraction and matching (SEM) pipeline. 1 Given a document, SEM extracts relevant skills and proficiency levels from using LLM prompting. 2 For each extracted skill, three candidates from ESCO taxonomy are selected using string matching and embedding similarities. 3 The extracted skill and taxonomy candidates are prompted to an LLM to find the best match.
• Figure 2: Illustration of the JCRec pipeline: In the offline phase, SEM is used to 1 extract skills required to take each course and skills provided by the courses, and to 2 extract skills required by each job posting skills. During the online phase, as a user uploads their CV, 1SEM extract their skills. 2 These skills are used to filter the set of courses they can enroll in. 3 From these, one course is recommended, aiming to maximize the increase in the number of jobs the user can apply for. 4 The user's profile is then updated with the skills acquired from the recommended course. 5 In the case of sequential recommendation, steps 2, 3, and 4 are repeated until $k$ courses are recommended.