NEO: A Tool for Taxonomy Enrichment with New Emerging Occupations

Giabelli, Anna; Malandri, Lorenzo; Mercorio, Fabio; Mezzanzanica, Mario; Seveso, Andrea

doi:10.1007/978-3-030-62466-8_35

Cited by 16 publications

(9 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We compare JAMES against a exhaustive list of ten baseline models relevant to the JTM task: KNN-based [38], Word2Vec-based [3], DeepCarotene [32], Node2Vec [14], GloVe [27], NEO [13], WoLMIS [2], BERT-based [30], Job2Vec [35], and Universal Sentence Encoder (USE) [5]. The details of baselines and our implementation settings are described in the Appendix A.…”

Section: Set-upmentioning

confidence: 99%

JAMES: Job Title Mapping with Multi-Aspect Embeddings and Reasoning

Yamashita¹,

Shen²,

Ekhtiari³

et al. 2022

Preprint

View full text Add to dashboard Cite

One of the most essential tasks needed for various downstream tasks in career analytics (e.g., career trajectory analysis, job mobility prediction, and job recommendation) is Job Title Mapping (JTM), where the goal is to map user-created (noisy and non-standard) job titles to predefined and standard job titles. However, solving JTM is domain-specific and non-trivial due to its inherent challenges:(1) user-created job titles are messy, (2) different job titles often overlap their job requirements, (3) job transition trajectories are inconsistent, and (4) the number of job titles in real world applications is large-scale. Toward this JTM problem, in this work, we propose a novel solution, named as JAMES, that constructs three unique embeddings of a target job title: topological, semantic, and syntactic embeddings, together with multi-aspect co-attention. In addition, we employ logical reasoning representations to collaboratively estimate similarities between messy job titles and standard job titles in the reasoning space. We conduct comprehensive experiments against ten competing models on the large-scale real-world dataset with more than 350,000 job titles. Our results show that JAMES significantly outperforms the best baseline by 10.06% in Precision@10 and by 17.52% in NDCG@10, respectively. Recently [20] built a 30,000 job title taxonomy on LinkedIn for a job understanding task. However, little is known about how the JTM task with the aforementioned challenges can be solved.Proposed Ideas. Toward these challenges, in this paper, we propose JAMES (Job title mApping with Multi-aspect Embeddings and rea Soning) to solve the JTM task. We use a large-scale and real-world career dataset with more than 350,000 job titles that a

show abstract

Section: Set-upmentioning

confidence: 99%

JAMES: Job Title Mapping with Multi-Aspect Embeddings and Reasoning

Yamashita¹,

Shen²,

Ekhtiari³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…NEO (Giabelli et al 2020b) This activity is crucial to allow economists and policy makers to observe up-to-date labour market dynamics using standard taxonomies as a lingua franca, overcoming linguistic boundaries (see, e.g. (Frey and Osborne 2017;Giabelli et al 2020a;Colombo, Mercorio, and Mezzanzanica 2019)).…”

Section: Overview Of Neomentioning

confidence: 99%

“…To select the best representing vectors, we rely on three distinct sub-tasks, that are the following: T1.1: train three different word embedding models (Word2Vec, GloVe, FastText), T1.2. construct measure of pairwise semantic similarity between taxonomic elements, namely Hierarchical Semantic Relatedness (HSR) (Giabelli et al 2020b). Compared with HSR, state-of-the-art metrics for semantic similarity (see Aouicha, Taieb, and Hamadou (2016) for a survey) suffer of two main drawbacks.…”

Section: Overview Of Neomentioning

confidence: 99%

See 1 more Smart Citation

NEO: A System for Identifying New Emerging Occupation from Job Ads

Giabelli

Malandri

Mercorio

et al. 2021

AAAI

Self Cite

View full text Add to dashboard Cite

We demonstrate NEO, a tool for automatically enriching the European Occupation and Skill Taxonomy (ESCO) with terms that represents new occupations extracted from million Online Job Advertisements (OJAs). NEO proposes (i) a novel metric that allows one to measure the semantic similarity between words in a taxonomy, and (ii) a set of measures that estimate the adherence of new terms to the most suited taxonomic concept, enabling the user to evaluate the suggestions. To test its effectiveness, NEO has been evaluated over 2M+ 2018 UK job ads, along with a user-study to confirm the usefulness of NEO in the taxonomy enrichment task.

show abstract

“…The workflow of skills2job, presented in Fig. 1, can be divided in five main steps: S.1 To extract linguistic patterns from OJVs, we train multiple embedding models through FastText, a library for representation learning which builds word embeddings considering sub-word information by representing each word as the sum of its character n-gram vectors; S.2 we compute measure of pairwise semantic similarity between taxonomic elements, namely HSS (developed in (Giabelli et al 2020b) and previously called Hierarchical Semantic Relatedness (HSR)).…”

Section: An Overview Of Skills2jobmentioning

confidence: 99%

Skills2Job: A Recommender System that Encodes Job Offer Embeddings on Graph Databases (Student Abstract)

Seveso

Giabelli

Malandri

et al. 2021

AAAI

Self Cite

View full text Add to dashboard Cite

We propose a recommender system that, starting from a set of users skills, identifies the most suitable jobs as they emerge from a large text of Online Job Vacancies (OJVs). To this aim, we process 2.5M+ OJVs posted in three different countries (United Kingdom, France and Germany), generating several embeddings and performing an intrinsic evaluation of their quality. Besides, we compute a measure of skill importance for each occupation in each country, the Revealed Comparative Advantage (rca). The best vector models, together with the rca, are used to feed a graph database, which will serve as the keystone for the recommender system. Finally, a user study of 10 validates the effectiveness of Skills2Job, both in terms of precision and nDGC.

show abstract

NEO: A Tool for Taxonomy Enrichment with New Emerging Occupations

Cited by 16 publications

References 27 publications

JAMES: Job Title Mapping with Multi-Aspect Embeddings and Reasoning

JAMES: Job Title Mapping with Multi-Aspect Embeddings and Reasoning

NEO: A System for Identifying New Emerging Occupation from Job Ads

Skills2Job: A Recommender System that Encodes Job Offer Embeddings on Graph Databases (Student Abstract)

Contact Info

Product

Resources

About