---
license: cc-by-4.0
language: en
tags:
- retrieval
- skill-extraction
- esco
- graded-relevance
configs:
  - config_name: queries
    data_files:
    - split: validation
      path: queries/validation.parquet
    - split: test
      path: queries/test.parquet
  - config_name: corpus
    data_files:
    - split: corpus
      path: corpus/corpus.parquet
  - config_name: qrels
    data_files:
    - split: validation
      path: qrels/validation.parquet
    - split: test
      path: qrels/test.parquet
---

# skill-extraction-tech-graded

Graded-relevance annotations for sentences from
[`TechWolf/skill-extraction-tech`](https://huggingface.co/datasets/TechWolf/skill-extraction-tech)
against the ESCO v1.1.0 skill taxonomy. Layout follows the
[BEIR](https://github.com/beir-cellar/beir) convention. 
The validation split carries full graded (0-4) relevance. A test split is also available with real but binary relevance; see the Test split section below.

## Configs

| config | rows | columns |
|---|---:|---|
| `queries` | 75 | `_id` (sentence id), `text` (sentence) |
| `corpus`  | 13,891 | `_id` (ESCO skill URI), `title` (English preferred label), `text` (English description), `esco_version` |
| `qrels`   | 1,041,825 | `query-id`, `corpus-id`, `score` (0-4) |

### Score scale

| score | volume | meaning |
|---:|---:|---|
| 0 | 1,018,085 | The skill is totally unrelated to the sentence.
| 1 | 22,923 | The skill's domain is correct. It's a plausible skill in a broader context, but not mentioned in this sentence.
| 2 | 339 | The skill could be recommended, but it's granularity makes it not core to the query.
| 3 | 309 | The skill is strongly relevant for this query, although it is more implied than explicitly demonstrated.
| 4 | 169 | The skill is explicitly demonstrated or requested by the query, and is therefor a clearly correct recommendation.

## Test split

A `test` split is now available (338 queries, 583 qrels rows).

Unlike the `validation` split, the test split is **not yet fully graded** (0-4). Its relevance labels are **real but binary**: `score = 1` marks a genuinely relevant target (derived from the public non-graded ground truths), and every pair not listed is implicit grade 0. The fine-grained 0-4 graded annotations for the test split are withheld during the ongoing RecSys-HR challenge (see [`WorkRB website`](https://techwolf-ai.github.io/workrb-site/challenges/recsys-hr-2026.html)) and will be released afterwards. 

## Attribution

This dataset uses the ESCO classification of the European Commission
(ESCO v1.1.0, <https://esco.ec.europa.eu>), licensed under
[CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). The ESCO content
has been extracted into a tabular subset (skill URI, English preferred label,
English description); no semantic modifications were made. The European
Commission is not responsible for any use of the data.