--- pretty_name: "Jobportal Data" tags: - webarena-pro - web-agents - benchmark-environment - text - tabular --- **Jobportal Data** provides job postings, company profiles, and resume-derived candidate inputs for WebArena-Pro’s Jobportal environment, a locally hosted recruitment application for evaluating web agents. It supports searching and filtering jobs, inspecting companies and candidates, and handling applications. Job and company records come directly from [LinkedIn Job Postings 2023–2024](https://www.kaggle.com/datasets/arshkon/linkedin-job-postings). Resume text comes from [Resume Classification Dataset for NLP](https://www.kaggle.com/datasets/hassnainzaidi/resume-classification-dataset-for-nlp). Kaggle is the direct distribution source; LinkedIn postings are upstream content for the jobs collection. We prepared the released content for local use in the environment: - Built a deterministic proportional sample by work type, with category enrichment for underrepresented portal categories. - Prepared compact CSV tables for jobs, companies, and classified resume profiles. - Mapped imported candidates to local benchmark identities and used resume categories to create relevant application fixtures in the application importer. | Statistic | Value | |---|---:| | Job postings | 10807 | | Companies | 5226 | | Resume profiles | 2484 | | Sampling seed | 20260710 | | Local release files before documentation | 6 | | Local payload size | 20,416,875 bytes (20.42 MB / 19.47 MiB) | Compressed UTF-8 CSV tables and the import manifest are under `seed_assets/`; selection and application-import code is under `preparation/`. These are complete import inputs; the release does not represent a separately inspected production database. The LinkedIn dataset is declared CC BY-SA 4.0. Resume-source terms are recorded as CC0/Public Domain in the project import manifest; that declaration has not been independently confirmed upstream. Source text may contain personal information and retains its applicable source rights. The release does not assign a blanket license to synthetic account/application fixtures. The application is maintained in the [WebArena-Pro repository](https://github.com/McGill-NLP/webarena-pro). ## Code and data licenses | Component | Recorded license or status | |---|---| | Preparation code in `preparation/` | No explicit license declaration was found in the WebArena-Pro source repository for these scripts. No new software license is assigned by this release. | | Data and content | LinkedIn Job Postings: CC BY-SA 4.0. Resume Classification Dataset: CC0/Public Domain as recorded in the project import metadata; that upstream declaration remains unverified. Synthetic account and application fixtures have no separately declared data license. |