Home/Office & Educational/Mathematical Word Problem Semantic Parsing and Question Answering Evaluation Dataset
Mathematical Word Problem Semantic Parsing and Question Answering Evaluation Dataset
#semantic parsing #quantitative relationship extraction #question answering #mathematical problem solving #educational assessment #quantitative reasoning
- 5000
- 0.3GB
- JSON
- CATL
MOBIUSI INC
Updated:2026-10-19
AI Analysis & Value Prop
This dataset covers mathematical word problems, their stated conditions, question targets, and reference answers, with structured details for quantities, units, relationships, operations, and implicit conditions. It includes normalized questions, question types, and structured answer targets to help assess how well models interpret natural-language problems and turn them into answerable questions. It is suitable for developing mathematical QA systems, studying semantic parsing, and evaluating models in the domain.
Dataset Insights
Sample Examples
{}
Technical Specifications
| Field | Type | Description |
|---|---|---|
| file_name | string | File name |
| operation | string | Identifies the primary mathematical operation inferred from the problem, such as addition, subtraction, multiplication, division, or comparison. |
| relations | array | Records quantity, comparison, or operational relationships parsed between entities in the problem. |
| quantities | array | Lists quantities, units, entities, and their semantic roles as identified from the problem conditions. |
| answer_target | object | Parses the reference answer into a structured target that can be calculated or validated. |
| question_type | string | Categorizes the question by its goal, such as finding a sum, difference, multiple, comparison, or remaining quantity. |
| problem_record | object | Contains the original text of the mathematical word problem, its question, and the reference answer. |
| unit_constraints | array | Specifies unit-consistency requirements for calculating or comparing quantities and expressing the answer. |
| implicit_conditions | array | Captures conditions inferred from common knowledge, grammar, or problem structure that are not fully stated but are relevant to solving the problem. |
| normalized_question | string | Rephrases the original question to make its meaning explicit, resolve references, and support calculation or direct answering. |
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 contain?
- It is organized around mathematical word problems and includes original problem records, normalized questions, question types, core operations, quantity entities and relations, implicit conditions, unit constraints, and structured answer targets.
- Which model capabilities can it evaluate?
- It can support evaluation of problem interpretation, reference resolution, quantity and unit recognition, relation extraction, operation selection, and answer-target parsing. It can also inform the development of mathematical QA and semantic parsing systems.
- How can the dataset be used for evaluation?
- Compare model outputs with the normalized questions, parsed relations, and structured answer targets, and report results by question type or parsing stage. Follow the dataset documentation to standardize answer formats, unit handling, and scoring rules.
- How can I access the dataset, and what are the licensing terms?
- Availability, supported formats, and permitted uses are specified on the product page and in the accompanying data license. Review the terms and applicable rights before commercial use or redistribution.
Can't find the data you need?
Post a request and let data providers reach out to you.
Get this Dataset
Verified for Enterprise UseCite this Work
@dataset{Mobiusi2026,
title={Mathematical Word Problem Semantic Parsing and Question Answering Evaluation Dataset},
author={MOBIUSI INC},
year={2026},
url={https://www.mobiusi.com/datasets/80caf89c4da9369edce8f899ac48c3dc?dataset_scene_cate_type=2},
urldate={2026-10-15},
keywords={mathematical word problem dataset,math QA evaluation,word problem semantic parsing,quantity relation extraction,math reasoning data,question normalization,AI in education,QA system evaluation},
version={1.0}
}Using this in research? Please cite us.
placeholder
placeholder
placeholder
placeholder
placeholder
placeholder
placeholder