JobBERT: Understanding Job Titles through Skills

Decorte, Jens-Joris; Hautte, Jeroen Van; Demeester, Thomas; Develder, Chris

doi:10.48550/arxiv.2109.09605

Cited by 6 publications

(9 citation statements)

References 6 publications

(12 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, Decorte et al [17] proposed a neural representation model for job titles, called JobBERT, by supplementing the pre-trained language model with co-occurrence information from skill labels extracted from job postings. The model leads to significant improvements over the use of generic sentence encoders for the title normalization work task, for which we publish a new evaluation benchmark.…”

Section: Related Workmentioning

confidence: 99%

Automated occupation coding with hierarchical features: a data-centric approach to classification with pre-trained language models

Safikhani

Avetisyan²,

Föste-Eggers³

et al. 2023

Discov Artif Intell

View full text Add to dashboard Cite

Occupation coding is the classification of information on occupation that is collected in the context of demographic variables. Occupation coding is an important, but a tedious task for researchers in social science and official statistics that calls for automation. Due to the complexity of the task, currently, researchers carry out hand-coding or computer-assisted coding. However, we argue that, with the rise of transformer-based language models, hand-coding can be displaced by models, such as BERT or GPT3. Hence, we compare these models with state-of-the-art encoding approaches, showing that language models have a clear advantage in Cohen’s kappa compared to related approaches, but also allow for flexible fine-grained coding of single digits. Taking into consideration the hierarchical structure of the occupational group, we also develop an approach that achieves better performance for the classification of different single digit combinations.

show abstract

Section: Related Workmentioning

confidence: 99%

Automated occupation coding with hierarchical features: a data-centric approach to classification with pre-trained language models

Safikhani

Avetisyan²,

Föste-Eggers³

et al. 2023

Discov Artif Intell

View full text Add to dashboard Cite

show abstract

“…In addition, they presented an encode-decode based fusion method to obtain a unified representation from the multi-view representations. JobBERT is the latest effort in this trajectory, where the authors of [39] presented a neural representation model for job titles, by augmenting a pre-trained language model with co-occurrence information from skill labels extracted from job vacancies.…”

Section: B Representation Learningmentioning

confidence: 99%

Knowledge Graphs in Education and Employability: A Survey on Applications and Techniques

2022

View full text Add to dashboard Cite

Studies on the relationship between education and employability are of paramount importance for policy makers, training institutions, companies and students. The availability of large scale data on the internet, such as online job ads, has been leveraged to better investigate this relationship. Recent studies have shown that representing this data using knowledge graphs can be useful in identifying the job market needs and establishing better assessment methods. Such studies also highlight the skill mismatch between education and the job market and help mitigate it. In this survey, we provide a comprehensive review of the applications of knowledge graphs in education and employability. We go through the knowledge graph based frameworks that have been proposed in previous work on different aspects of education and employability. The survey introduces new taxonomies for these applications and their methodologies, and provides a thorough outlook on promising research directions. We also present a use case to illustrate how knowledge graphs can be used to investigate the relationship between education and the job market.

show abstract

“…Although job titles generally follow a certain structure [1], they are often freely structured and could contain some degree of bias. There was some work carried out aiming at normalizing job titles and relating their free form to the standardized job title form [1,4], [5].…”

Section: Introductionmentioning

confidence: 99%

“…However [4] points out that only 65 % of job titles could be linked to their corresponding normalized taxonomy without ambiguity. This leaves 35 % of Job Ads for which it is of utmost importance to analyze requirements specified in the job description section to map the job to the taxonomy.…”

Section: Introductionmentioning

confidence: 99%

“…To bypass the need for a large hand-labeled dataset, we, following authors of [4,5] make use of job ads and work experience sections extracted from anonymized resumes to learn the effective representation of titles and skills mentioned in requirements sections of vacancies. To exploit this signal distant supervision settings (in which there is no need for explicitly human-annotated data) are applied in which our semantic similarity model learns to understand the meaning of requirements present in the Job Ad and map them to corresponding job titles.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

VacancySBERT: the approach for representation of titles and skillsfor semantic similarity search in the recruitment domain

Bocharova¹,

Malakhov²,

Mezhuyev³

2023

AAIT

View full text Add to dashboard Cite

The paper focuses on deep learning semantic search algorithms applied in the HR domain. The aim of the article is developing a novel approach to training a Siamese network to link the skills mentioned in the job ad with the title. It has been shown that the title normalization process can be based either on classification or similarity comparison approaches. While classification algorithms strive to classify a sample into predefined set of categories, similarity search algorithms take a more flexible approach, since they are designed to find samples that are similar to a given query sample, without requiring pre-defined classes and labels. In this article semantic similarity search to find candidates for title normalization has been used. A pre-trained language model has been adapted while teaching it to match titles and skills based on co-occurrence information. For the purpose of this research fifty billion titledescriptions pairs had been collected for training the model and thirty three thousand title-description-normalized title triplets, where normalized job title was picked up manually by job ad creator for testing purposes. As baselines FastText, BERT, SentenceBert and JobBert have been used. As a metric of the accuracy of the designed algorithm is Recall in top one, five and ten model’s suggestions. It has been shown that the novel training objective lets it achieve significant improvement in comparison to other generic and specific text encoders. Two settings with treating titles as standalone strings, and with included skills as additional features during inference have been used and the results have been compared in this article. Improvements by 10 % and 21.5 % have been achieved using VacancySBERT and VacancySBERT (with skills) respectively. The benchmark has been developed as open-source to foster further research in the area

show abstract

JobBERT: Understanding Job Titles through Skills

Cited by 6 publications

References 6 publications

Automated occupation coding with hierarchical features: a data-centric approach to classification with pre-trained language models

Automated occupation coding with hierarchical features: a data-centric approach to classification with pre-trained language models

Knowledge Graphs in Education and Employability: A Survey on Applications and Techniques

VacancySBERT: the approach for representation of titles and skillsfor semantic similarity search in the recruitment domain

Contact Info

Product

Resources

About