Retrieving Skills from Job Descriptions: A Language Model Based Extreme Multi-label Classification Framework

Abstract

We introduce a deep learning model to learn the set of enumerated job skills associated with a job description. In our analysis of a large-scale government job portal mycareersfuture.sg, we observe that as much as 65% of job descriptions miss describing a significant number of relevant skills. Our model addresses this task from the perspective of an extreme multi-label classification (XMLC) problem, where descriptions are the evidence for the binary relevance of thousands of individual skills. Building upon the current state-of-the-art language modeling approaches such as BERT, we show our XMLC method improves on an existing baseline solution by over 9% and 7% absolute improvements in terms of recall and normalized discounted cumulative gain.We further show that our approach effectively addresses the missing skills problem, and helps in recovering relevant skills that were missed out in the job postings by taking into account the structured semantic representation of skills and their co-occurrences through a Correlation Aware Bootstrapping process. To facilitate future research and replication of our work, we have made the dataset and the implementation of our model publicly available.

SkillSpan: Hard and Soft Skill Extraction from English Job Postings

“…As can be seen in Table 1, many works do not release their data (apart from Sayfullina et al, 2018 andBhola et al, 2020) and none release their annotation guidelines. In addition, none of the previous studies approach SE as a span-level extraction task with state-of-the-art language models, nor did 1 https://github.com/kris927b/SkillSpan they release a dataset of this magnitude with manually annotated (long) spans of competences by domain experts.…”

“…Previous work in SE shows promising progress, but is halted by a lack of available datasets and annotation guidelines. Two out of 14 studies release their dataset, which limit themselves to crowd-sourced labels (Sayfullina et al, 2018) or annotations from a predefined list of skills on the document-level (Bhola et al, 2020). Additionally, none of the 14 previously mentioned studies release their annotation guidelines, which obscures the meaning of a competence.…”

Proceedings of the 1st Workshop on NLP for Positive Impact

“…In our context, for example, the "labels" that curricular and job content share are skills. Due to the sheer volume of possible skills, this becomes an extreme multilabel classification task (XMC) as recently pointed out by Bhola et al (2020). In their closely relevant work, BERT models are employed to learn an embedding for a job description and XMC models then classify each embedding into a subset of skills over a large pool of predetermined skills set.…”

“…Therefore, computing Equation (1) translates into the task of predicting skills from the content of syllabi. There can be multiple NLP approaches for this task, and here we present an example inspired by Bhola et al (2020) who frame skill identification as a multilabel classification problem. Specifically, we describe a BERT-LSTM architecture as illustrated in Figure 1.…”

“…The top part of Figure 1 is a multilabel classifier constructed as a feed-forward neural network where the prediction targets are skill labels. Additional tweaks such as Correlation Aware Bootstrapping (Bhola et al, 2020) can be simply added for the sake of performance.…”