Add dataset card
Browse files
README.md
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pretty_name: SkillSpace Data
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
license: mit
|
| 6 |
+
task_categories:
|
| 7 |
+
- text-retrieval
|
| 8 |
+
- text-classification
|
| 9 |
+
tags:
|
| 10 |
+
- resumes
|
| 11 |
+
- jobs
|
| 12 |
+
- semantic-search
|
| 13 |
+
- retrieval
|
| 14 |
+
- careers
|
| 15 |
+
size_categories:
|
| 16 |
+
- n<1K
|
| 17 |
+
---
|
| 18 |
+
|
| 19 |
+
# SkillSpace Data
|
| 20 |
+
|
| 21 |
+
This dataset repo stores the cleaned Phase 1 artifacts used by SkillSpace.
|
| 22 |
+
|
| 23 |
+
## Contents
|
| 24 |
+
|
| 25 |
+
- `processed/jobs_hf_v1.parquet`
|
| 26 |
+
- `processed/resumes_hf_v1.parquet`
|
| 27 |
+
- `sample_data/skills_seed.csv`
|
| 28 |
+
- `sample_data/jobs_sample.csv`
|
| 29 |
+
- `sample_data/resumes_sample.csv`
|
| 30 |
+
|
| 31 |
+
## Purpose
|
| 32 |
+
|
| 33 |
+
These artifacts support:
|
| 34 |
+
|
| 35 |
+
- semantic resume-to-job retrieval
|
| 36 |
+
- deterministic skill-gap reasoning
|
| 37 |
+
- Gradio demo evaluation and iteration
|
| 38 |
+
|
| 39 |
+
## Provenance
|
| 40 |
+
|
| 41 |
+
The cleaned artifacts were derived locally from public Hugging Face source datasets and normalized into a project-owned schema for Phase 1 development.
|
| 42 |
+
|
| 43 |
+
## Notes
|
| 44 |
+
|
| 45 |
+
- This is an evolving project dataset, not a final benchmark release.
|
| 46 |
+
- Future versions may expand rows, improve normalization, and add evaluation splits.
|