Hugging Face's logo

Datasets:
Aixr
/
Aixr-1M

Tasks:
Question Answering
Modalities:
Text
Formats:
parquet
Languages:
Turkish
English
Size:
100K - 1M
Tags:
code
chemistry
medical
biology
Libraries:
Datasets
pandas
License:
Dataset card Viewer Files Community
1

You need to agree to share your contact information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this dataset content.

Multilingual and Modular Question-Answer Dataset for AI

Introduction

In the rapidly evolving field of artificial intelligence, the availability of robust and diverse datasets is paramount for training effective question-answering (QA) systems. This document introduces a comprehensive multilingual and modular QA dataset designed to cater to both English and Turkish languages. The dataset encompasses a wide range of mathematical operations and sentence structures, making it a versatile resource for developing and enhancing AI-driven QA models.

Dataset Overview

The dataset is meticulously crafted to support bilingual (English and Turkish) question-answering tasks. It integrates mathematical operations seamlessly with natural language processing, ensuring that AI models can handle both numerical and linguistic queries with equal proficiency.

Key Features

  • Multilingual Support: Covers both English and Turkish languages, enabling the development of QA systems that cater to a diverse user base.
  • Mathematical Operations: Includes a variety of mathematical questions ranging from basic arithmetic to complex calculations, fostering the model's ability to process numerical data.
  • Rich Sentence Structures: Contains diverse sentence types and complexities in both languages, enhancing the model's understanding of different linguistic constructs.
  • Modular Design: Structured in a modular fashion, allowing for easy scalability and adaptability to various use cases and extensions.

Dataset Composition

Languages

  • English: Comprehensive coverage of English language questions and answers, suitable for native speakers and learners alike.
  • Turkish: Equally thorough in Turkish, ensuring that the dataset is valuable for Turkish-speaking users and applications.

Mathematical Content

The dataset includes questions that involve:

  • Basic Arithmetic: Addition, subtraction, multiplication, and division.
  • Algebra: Solving for variables, equations, and inequalities.
  • Geometry: Questions related to shapes, areas, and volumes.
  • Advanced Mathematics: Topics such as calculus, probability, and statistics.

Sentence Structures

  • Simple Sentences: Straightforward questions that are easy to parse.
  • Complex Sentences: Involving multiple clauses and intricate syntax.
  • Contextual Questions: Questions that require understanding of context and nuanced language use.

Modular Structure

The dataset is organized into distinct modules, each focusing on a specific aspect of question-answering. This modularity facilitates targeted training and fine-tuning of AI models, allowing developers to focus on particular areas of interest or need.

Modules Include:

  1. Language Module: Separate sub-modules for English and Turkish, each containing language-specific questions and answers.
  2. Mathematics Module: Divided into subcategories based on mathematical topics, ensuring focused training on numerical operations.
  3. Mixed Module: Combines linguistic and mathematical questions to train models on handling multi-faceted queries.
  4. Contextual Module: Contains questions that require contextual understanding, enhancing the model's ability to interpret and respond accurately.

Usage Guidelines

This dataset is designed to be flexible and user-friendly, accommodating various use cases such as:

  • Training QA Models: Ideal for training machine learning models in both English and Turkish.
  • Benchmarking: Useful for evaluating the performance of existing QA systems across multiple languages and domains.
  • Research and Development: Supports academic and industrial research aimed at improving multilingual and mathematical question-answering capabilities.

Access and Licensing

The dataset is available under an open-source license, encouraging collaboration and innovation within the AI community. Users are free to download, modify, and distribute the dataset in accordance with the licensing terms.

Example Entries

Below are sample entries from the dataset to illustrate its structure and content:

English

Question: What is the sum of 45 and 78?

Answer: 123

Question: Solve for x in the equation 2x + 5 = 17.

Answer: x = 6

Question: Who wrote the play "Hamlet"?

Answer: William Shakespeare

Turkish

Question: 45 ile 78'in toplamı nedir?

Answer: 123

Question: 2x + 5 = 17 denklemini çözün.

Answer: x = 6

Question: "Hamlet" oyununu kim yazdı?

Answer: William Shakespeare

Conclusion

This multilingual and modular question-answer dataset serves as a robust foundation for developing sophisticated AI-driven QA systems. By encompassing both English and Turkish languages, integrating a wide range of mathematical operations, and adopting a modular structure, the dataset ensures versatility and scalability. Whether for training, benchmarking, or research, this dataset is an invaluable resource for advancing the capabilities of artificial intelligence in understanding and responding to diverse and complex queries.

Contact and Support

For further information, support, or contributions, please contact the dataset maintainers at meforgers@meforgers.com

Downloads last month
10