CoolFace
Datasetpublic

Yazeed-Kamel/hashem-dataset

Hashem AI – Jordanian Legal FAQ Dataset Overview This dataset contains a curated collection of frequently asked questions (FAQ) and authoritative legal answers related to Jordanian laws and regulations.It is designed to support Arabic-language legal question answering, retrieval-augmented generation (RAG), and legal knowledge systems. The dataset focuses on providing clear, neutral, and legally accurate explanations based strictly on official Jordanian legislation… See the full description on the dataset page: https://huggingface.co/datasets/Yazeed-Kamel/hashem-dataset.

sourceHugging Faceapache-2.0updated 8mo agoView on Hugging Face
2likes38downloads
Dataset Card

Hashem AI – Jordanian Legal FAQ Dataset

Overview

This dataset contains a curated collection of frequently asked questions (FAQ) and authoritative legal answers related to Jordanian laws and regulations. It is designed to support Arabic-language legal question answering, retrieval-augmented generation (RAG), and legal knowledge systems.

The dataset focuses on providing clear, neutral, and legally accurate explanations based strictly on official Jordanian legislation and regulations.


Legal Domains Covered

The dataset includes content from the following legal areas:

  • —Jordanian Labor Law
  • —Social Security Law
  • —Work Permits and Employment Regulations
  • —Workers’ Rights and Obligations
  • —Overtime, Working Hours, and Leave
  • —Employment Contracts and Termination
  • —Refugee Employment and Legal Status
  • —Ministry of Labor Procedures and Contact Channels

Dataset Structure

Each record in the dataset contains:

FieldDescription
idUnique identifier for the FAQ entry
questionLegal question in Arabic
answerOfficial, legally grounded answer in Arabic
categoryHigh-level legal category
tagsKeywords for search and retrieval
is_deletedLogical deletion flag
is_activeAvailability flag
created_atRecord creation timestamp

Intended Use Cases

This dataset is suitable for:

  • —Legal Question Answering (QA) systems
  • —Retrieval-Augmented Generation (RAG) pipelines
  • —Chatbots and virtual legal assistants
  • —Semantic search and knowledge retrieval
  • —Legal NLP research in Arabic
  • —Training and evaluating LLMs on domain-specific legal content

Data Principles

  • —Authoritative: Answers are derived from official Jordanian laws, regulations, and government sources.
  • —Neutral and Non-Advisory: The dataset does not provide personalized legal advice.
  • —Clear Language: Legal explanations are written in clear and accessible Arabic.
  • —Structured: Optimized for indexing, retrieval, and embedding-based search.

Limitations

  • —This dataset does not replace professional legal consultation.
  • —It does not handle case-specific or personalized legal scenarios.
  • —Legal texts may change over time; users should verify against the latest official legislation.

Ethical and Legal Disclaimer

This dataset is provided for informational and educational purposes only. It does not constitute legal advice, nor does it replace qualified legal professionals or official government channels.


License

This dataset is released under the Apache License 2.0, allowing free use, modification, and redistribution with proper attribution.


Attribution

Developed as part of the Hashem AI initiative, focused on improving access to Jordanian legal information through AI-powered systems.


Contact

For technical issues, dataset improvements, or corrections, please open an issue in the repository.

For official legal inquiries or complaints, refer to the relevant Jordanian government authorities.