Learning Representations for Soft Skill Matching

TL;DR

This paper presents a neural network-based method for accurately detecting and disambiguating soft skill phrases in job ads by modeling context, significantly improving recall while maintaining high precision.

Contribution

It introduces a context-aware phrase matching approach with soft skill tagging and compares multiple neural models, achieving high recall in soft skill detection.

Findings

LSTM with soft skill tagging achieved 83.92% recall at 95% precision.

Context modeling improves soft skill phrase disambiguation.

Neural network approaches outperform naive matching methods.

Abstract

Employers actively look for talents having not only specific hard skills but also various soft skills. To analyze the soft skill demands on the job market, it is important to be able to detect soft skill phrases from job advertisements automatically. However, a naive matching of soft skill phrases can lead to false positive matches when a soft skill phrase, such as friendly, is used to describe a company, a team, or another entity, rather than a desired candidate. In this paper, we propose a phrase-matching-based approach which differentiates between soft skill phrases referring to a candidate vs. something else. The disambiguation is formulated as a binary text classification problem where the prediction is made for the potential soft skill based on the context where it occurs. To inform the model about the soft skill for which the prediction is made, we develop several approaches, including soft skill masking and soft skill tagging. We compare several neural network based approaches, including CNN, LSTM and Hierarchical Attention Model. The proposed tagging-based input representation using LSTM achieved the highest recall of 83.92% on the job dataset when fixing a precision to 95%.