Inductive Graph Neural Network for Job-SkillFramework Analysis

The analysis of skills and their relationship to different occupations is anarea of special attention within human capital management processes. Nowadays,job specialization has made this increasingly important. In this paper, we addresstwo main tasks: the retrieval of similar jobs and the retrieval of skills related to agiven job. We develop a system that combines the encoding of textual informationwith a graph neural network, thus mitigating the limitations of a system that relieson either of these separately. We present experiments that show that the proposedsystem is able to take advantage of both the encoded textual information, andthe structured relationships between job titles and skills represented by the graph.We also show the robustness of the proposed model in modeling unseen entitiesby evaluating the model's performance in simulated cold-recommendation scenarioswhere a percentage of the skills under study are eliminated during training.