TalentCLEF 2025 corpus: Skill and Job Title Intelligence for Human Capital Management
<blockquote>
<p><strong>🚨 Current Status: Submission of Working Notes</strong></p>
</blockquote>
<p><strong>If you use any data from this repository, please cite our scientific paper instead of the Zenodo repo: </strong></p>
<pre><code>@inproceedings{gasco2025overview,
title={{Overview of the TalentCLEF 2025: Skill and Job Title Intelligence for Human Capital Management}},
author={Gasco, Luis and Fabregat, Hermenegildo and Garc\'{\i}a-Sardi{\~n}a, Laura and Estrella, Paula and Deniz, Daniel and Rodrigo, \'{A}lvaro and Zbib, Rabih},
booktitle={{International Conference of the Cross-Language Evaluation Forum for European Languages}},
year={2025},
publisher={Springer}
}</code></pre>
<p> </p>
<h2><strong><a href="https://talentclef.github.io/talentclef/" target="_blank" rel="noopener">TalentCLEF2025</a> corpus - </strong>Task B Test set release</h2>
<h3><strong>Introduction:</strong></h3>
<p>The first edition of TalentCLEF aims to develop and evaluate models designed to facilitate three essential tasks:</p>
<ol>
<li>Finding/ranking candidates for job positions based on their experience and professional skills.</li>
<li>Implementing upskilling and reskilling strategies that promote the coninuous development of workers</li>
<li>Detecting emerging skills and skills gaps of importance in organizations.</li>
</ol>
<p>With that aim, the task is divided into two tasks: </p>
<ul>
<li><strong>Task A - Multilingual Job Title Matching</strong>. This task involves developing systems to identify and rank the job titles most similar to a given one by generating a ranked list of similar titles from a specified knowledge base for each job title in a provided test set.</li>
<li><strong>Task B - Job Title-Based Skill Prediction. </strong>Task B requires developing systems that can retrieve relevant skills associated with a specified job title.</li>
</ul>
<div>
<p>This data repository contains the data for these two tasks. The data is being released progressively according to the<a href="https://talentclef.github.io/talentclef/docs/talentclef-2025/schedule/"> task schedule</a>.</p>
<p>The task evaluation takes place on Codabench (<a href="https://www.codabench.org/competitions/5842/">Task A</a> and <a href="https://www.codabench.org/competitions/7059/">Task B</a>). Participants must register for the competition through <a href="https://clef2025-labs-registration.dei.unipd.it/">CLEF Lab Registration Page</a> to be part of the evaluation campaign.</p>
<p> </p>
</div>
<h3><strong>File structure: </strong></h3>
<div>
<div>
<div>
<div>
<blockquote>
<p>For a detailed description of the data structure, you can refer to the <a href="https://talentclef.github.io/talentclef/docs/talentclef-2025/data/description_corpus/">TalentCLEF2025 data description page,</a> where it is thoroughly explained.</p>
</blockquote>
</div>
</div>
</div>
</div>
<p>The files is organized into two <code>*.zip</code> files, <code>TaskA.zip</code> and <code>TaskB.zip</code>, each containing training, validation and test folders to support different stages of model development. So far, only the training set for both tasks has been released, but in future releases, as the tasks progress, additional data will be added to the different subfolders for each task.</p>
<p><strong>TaskA</strong> includes language-specific subfolders within the training and validation directories, covering English, Spanish, German, and Chinese job title data. The training folders for TaskA contain language-specific .tsv files for each respective language. Validation folders include three essential files—queries, corpus_elements, and q_rels—for evaluating model relevance to search queries. TaskA’s test folder has queries and corpus_elements files for testing retrieval.<br><br></p>
<pre><code>TaskA/
│
├── training/
│ ├── english/
│ │ └── taskA_training_en.tsv
│ ├── spanish/
│ │ └── taskA_training_es.tsv
│ └── german/
│ └── taskA_training_de.tsv
│
├── validation/
│ ├── english/
│ │ ├── queries
│ │ ├── corpus_elements
│ │ └── qrels
│ ├── spanish/
│ ├── german/
│ └── chinese/
│
└── test/
├── english/
│ ├── queries
│ └── corpus_elements
├── spanish/
├── german/
└── chinese/
</code></pre>
<p><strong>TaskB</strong> follows a similar structure but without language-specific subfolders, providing general .tsv files for training, validation, and testing. This consistent file organization enables efficient data access and structured updates as new data versions are published.</p>
<pre><code>TaskB/
│
├── training/
│ ├── job2skill.tsv
│ ├── jobid2terms.json
│ └── skillid2terms.json<br>│
├── validation/
│ ├── queries
│ ├── corpus_elements
│ └── qrels
│
└── test/
├── queries
└── corpus_elements
</code></pre>
<p><strong>Tutorials:</strong></p>
<table>
<tbody>
<tr>
<td><strong>Notebook</strong></td>
<td>Link</td>
</tr>
<tr>
<td>Data Download and Load using Python </td>
<td><a href="https://colab.research.google.com/github/TalentCLEF/talentclef_tutorials/blob/main/talentclef2025/TalentCLEF_data_tutorial.ipynb" target="_blank" rel="noopener">Link to Colab</a></td>
</tr>
<tr>
<td>Task A - Prepare submission file and run evaluation</td>
<td><a href="https://colab.research.google.com/github/TalentCLEF/talentclef_tutorials/blob/main/talentclef2025/TalentCLEF_submission_creation_tutorial.ipynb" target="_blank" rel="noopener">Link to Colab</a></td>
</tr>
<tr>
<td>Task A - Development set Baseline generation</td>
<td><a href="https://colab.research.google.com/github/TalentCLEF/talentclef_tutorials/blob/main/talentclef2025/TalentCLEF_TaskA_DevSet_Baseline.ipynb">Link to Colab</a></td>
</tr>
<tr>
<td>Task B - Prepare submission file and run evaluation</td>
<td><a href="https://colab.research.google.com/github/TalentCLEF/talentclef_tutorials/blob/main/talentclef2025/TalentCLEF_TaskB_submission_creation_tutorial.ipynb" target="_blank" rel="noopener">Link to Colab</a></td>
</tr>
</tbody>
</table>
<p><strong>Resources: </strong></p>
<ul>
<li><a href="https://talentclef.github.io/talentclef/docs/" target="_blank" rel="noopener">Web</a></li>
<li><a href="https://clef2025-labs-registration.dei.unipd.it/">CLEF Lab Registration Page</a></li>
<li><a href="https://www.codabench.org/competitions/5842/">Codabench Task A</a></li>
<li><a href="https://www.codabench.org/competitions/7059/">Codabench Task B</a></li>
<li><a href="https://talentclef.github.io/talentclef/docs/talentclef-2025/data/additional_resources/">Additional Resources</a></li>
</ul>
opencc-by-4.0Oct 2024View details →
