Arabic Morphology and Weak Verb Question Answering Dataset

#question answering #morphological analysis #grammatical reasoning #Arabic morphology education #Arabic root analysis #weak verb conjugation

5000
0.4GB
JSON
CATL
MOBIUSI INC

Updated:2026-10-19

AI Analysis & Value Prop

Built around web text on Arabic morphology, this dataset covers fine-grained topics including roots, verb forms, derived forms, weak verb changes, and noun formation. It pairs questions and answers with source evidence and organizes examples by question type and morphology topic. The data supports specialized question answering, source-grounded response training, and evaluation of models' understanding of Arabic morphological rules.

Dataset Insights

Sample Examples

{}

Technical Specifications

Field	Type	Description
file_name	string	File name
answer	string	An accurate response grounded in the source text, which may explain roots, verb forms, derived forms, weak verb changes, or noun formation.
question	string	An Arabic morphology question generated from the source text; its content must not go beyond the supplied material.
source_text	string	Original Arabic morphology web content or cleaned web text used to generate the question-answer pair.
question_type	string	The question category, such as concept explanation, form identification, inflection, form comparison, or rule application.
source_evidence	array	Source text excerpts supporting the question and answer, such as short quotations or key terms.
knowledge_points	array	Fine-grained Arabic morphology knowledge points extracted from the question-answer pair.
morphology_topic	string	The main Arabic morphology topic addressed, such as roots, verb forms, derived forms, weak verbs, or noun formation.
is_source_grounded	boolean	Indicates whether the question and answer are fully supported by the source text; set to false when the source does not support them.
morphological_analysis	object	A structured summary of relevant roots, verb forms, derived forms, weak verb changes, and noun formation; inapplicable attributes may be omitted.

Compliance Statement

Authorization Type	Proprietary - Commercial AI Training License (No Redistribution)
Commercial Use	Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization	No PII, no real company names, simulated scenarios follow industry standards
Compliance System	Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs

Frequently Asked Questions

What does the dataset cover?: The dataset focuses on Arabic morphology in web text, including roots, verb forms, derived forms, weak verb changes, and noun formation. Records may include source text, questions, answers, evidence, question types, topics, and structured morphological analysis.

How are answers tied to their sources?: Questions and answers are based on the supplied source text. Evidence excerpts are included, along with a source-grounding indicator showing whether the content is supported by the source, enabling verification and grounded-answer training.

Which training and evaluation tasks does it support?: It is suitable for Arabic closed-domain question answering, morphological knowledge understanding, weak verb rule questions, source-grounded responses, and question type classification. It can support domain-specific supervised fine-tuning and model evaluation.

What are the dataset size, access method, and license?: The supplied information does not specify the record count, access method, or license terms. Check the product page for version details, delivery instructions, and the applicable license, and confirm the permitted scope before use.

Can't find the data you need?

Post a request and let data providers reach out to you.

Get this Dataset

Verified for Enterprise Use

Cite this Work

@dataset{Mobiusi2026,
  title={Arabic Morphology and Weak Verb Question Answering Dataset},
  author={MOBIUSI INC},
  year={2026},
  url={https://www.mobiusi.com/datasets/00677fd5956c77958218bca96266ed49},
  urldate={2026-10-15},
  keywords={Arabic morphology dataset,Arabic question answering dataset,Arabic morphology,Arabic weak verbs,Arabic roots,Arabic verb forms,Arabic NLP,source-grounded question answering},
  version={1.0}
}

Using this in research? Please cite us.

placeholder

Products

Scene

Domain

Modality

Task