---
license: cc-by-4.0
task_categories:
- text-generation
- text-classification
- question-answering
language:
- en
tags:
- scientific-papers
- academic-research
- hindawi
- biomedical
- research-articles
- filtered
- medicine
- clinical
- math
- engineering
- physics
- chemistry
- biology
- computer-science
- earth-sciences
- agricultural-sciences
- social-sciences
- business
- other
- geosciences
- veterinary
- cc-by-compatible
size_categories:
- 100K<n<1M
source_datasets:
- original
---

# Hindawi Academic Papers Dataset (CC BY 4.0 Compatible)

## Dataset Description

This dataset contains 299,316 academic research papers from Hindawi Publishing Corporation, carefully filtered to include only papers with licenses compatible with CC BY 4.0. The dataset includes comprehensive metadata for each paper including titles, authors, journal information, publication years, DOIs, and full-text content.

### Dataset Summary

- **Total Papers**: 299,316 (filtered from 299,363 original papers)
- **Source**: Hindawi Publishing Corporation
- **Format**: Structured dataset with metadata and full-text content
- **Language**: English
- **License**: CC BY 4.0 (all constituent papers are CC BY 4.0 compatible)
- **Time Period**: 2007 - 2023
- **Filtering**: Removed 47 papers with incompatible licenses (Unknown, BY-NC)

### Key Statistics

#### Content Statistics
- **Average content length**: 42,866 characters
- **Average authors per paper**: 4.7
- **Shortest paper**: 51 characters
- **Longest paper**: 939,992 characters

#### Temporal Distribution
- **Publication years**: 2007 - 2023
- **Papers with year information**: 299,192 (100.0%)

#### Data Completeness
- **Title**: 299,316 papers (100.0%)
- **Authors**: 299,298 papers (100.0%)
- **Journal**: 299,316 papers (100.0%)
- **Year**: 299,192 papers (100.0%)
- **Publisher**: 299,316 papers (100.0%)
- **License**: 299,316 papers (100.0%)
- **Doi**: 299,316 papers (100.0%)
- **Source File**: 299,316 papers (100.0%)

#### Top Journals (by number of papers)
- **BioMed Research International**: 23,967 papers (8.0%)
- **Mathematical Problems in Engineering**: 20,254 papers (6.8%)
- **Evidence-Based Complementary and Alternative Medicine**: 12,448 papers (4.2%)
- **The Scientific World Journal**: 7,923 papers (2.6%)
- **Oxidative Medicine and Cellular Longevity**: 7,154 papers (2.4%)
- **Journal of Nanomaterials**: 5,623 papers (1.9%)
- **Advances in Materials Science and Engineering**: 5,515 papers (1.8%)
- **Complexity**: 5,471 papers (1.8%)
- **Wireless Communications and Mobile Computing**: 5,308 papers (1.8%)
- **Computational Intelligence and Neuroscience**: 5,254 papers (1.8%)

#### Top Publishers (by number of papers)
- **Hindawi**: 161,322 papers (53.9%)
- **Hindawi Publishing Corporation**: 130,378 papers (43.6%)
- **International Scholarly Research Network**: 3,821 papers (1.3%)
- **SAGE-Hindawi Access to Research**: 2,430 papers (0.8%)
- **The Scientific World Journal**: 1,138 papers (0.4%)
- **Scholarly Research Exchange**: 110 papers (0.0%)
- **TheScientificWorldJOURNAL**: 66 papers (0.0%)
- **Hindawi Access to Research**: 5 papers (0.0%)
- **Case Reports in Anesthesiology**: 2 papers (0.0%)
- **Scientifica**: 2 papers (0.0%)

#### Categories
Number of categories: 13

All categories (sorted by count):
 1. Medical & Health Sciences: 150,633 papers (50.3%)
 2. Engineering & Technology: 87,251 papers (29.2%)
 3. Mathematical Sciences: 24,087 papers (8.0%)
 4. Chemistry and Chemical Sciences: 8,794 papers (2.9%)
 5. Biological Sciences: 7,064 papers (2.4%)
 6. Physical Sciences: 5,259 papers (1.8%)
 7. Computer Science: 5,178 papers (1.7%)
 8. Earth and Environmental Sciences: 3,396 papers (1.1%)
 9. Agricultural Sciences: 3,361 papers (1.1%)
10. Social Sciences & Business: 3,336 papers (1.1%)
11. Other: 932 papers (0.3%)
12. Geosciences: 21 papers (0.0%)
13. Veterinary Sciences: 4 papers (0.0%)

![Categories](image.png)

## Dataset Structure

### Data Fields

- `text`: Original markdown text of the paper
- `filename`: Original filename of the source file  
- `file_size`: Size of the original file in bytes
- `title`: Extracted paper title
- `category`: Category based on journal name
- `authors`: Semicolon-separated list of authors
- `journal`: Journal name and publication year
- `year`: Extracted publication year
- `publisher`: Publisher name
- `license`: License name (all CC BY 4.0 compatible)
- `license_url`: Original license URL (all CC BY 4.0 compatible)
- `doi`: Digital Object Identifier
- `source_file`: Source XML file reference
- `content`: Main text content (excluding metadata)

### Data Splits

This dataset does not have predefined splits. Users should create their own train/validation/test splits as needed for their specific use case.

## Dataset Creation

### Source Data

The data originates from Hindawi Publishing Corporation's collection of academic papers, originally in XML format and converted to markdown. Hindawi is an open-access publisher focusing on scientific, technical, and medical research.

### Data Processing and Filtering

1. **Initial Format**: Papers were originally in XML format
2. **Conversion**: Converted to markdown format
3. **Parsing**: Metadata extracted using regex patterns
4. **License Filtering**: **IMPORTANT** - Removed 47 papers with:
   - Unknown licenses (46 papers)
   - Non-commercial restrictions (CC BY-NC) (1 paper)
5. **Final Dataset**: Only papers with CC BY 4.0 compatible licenses retained

### Filtering Criteria

Papers were kept if they had licenses compatible with CC BY 4.0:
- ✅ **Kept**: CC BY 4.0, CC BY 3.0, CC BY 2.0, CC BY-SA 4.0, CC BY-SA 3.0, CC0 1.0, MIT, Apache 2.0, BSD
- ❌ **Removed**: CC BY-NC (any version), CC BY-ND, Unknown licenses, Custom licenses

### Personal and Sensitive Information

This dataset contains academic papers that are publicly available. Author names and affiliations are included as part of the academic publication metadata. No personal sensitive information beyond standard academic attribution is included.

## Considerations for Using the Data

### Social Impact of Dataset

This dataset can be used for:
- Academic research and literature analysis
- Training models for scientific text understanding
- Citation analysis and bibliometric studies
- Content analysis of academic publications
- Commercial applications (all papers are CC BY 4.0 compatible)

### Discussion of Biases

- **Geographic bias**: Papers may be skewed toward certain regions/institutions that publish with Hindawi
- **Temporal bias**: Coverage may vary across different time periods
- **Subject bias**: Hindawi's focus areas may not represent all academic disciplines equally
- **Language bias**: Dataset is primarily in English
- **License bias**: Dataset now excludes papers with non-commercial restrictions, potentially affecting content diversity

### Other Known Limitations

- Some metadata fields may be incomplete due to parsing limitations
- Original formatting and figures are not preserved in the text content
- Mathematical formulas may not be properly formatted in plain text
- 47 papers were removed due to license compatibility requirements

## Additional Information

### Licensing Information

This dataset is released under **CC BY 4.0** license. All constituent papers have been verified to have CC BY 4.0 compatible licenses, making this dataset safe for commercial use and redistribution.

**License Compatibility Verification**: Every paper in this dataset has been automatically verified to have a license compatible with CC BY 4.0. Papers with unknown, non-commercial, or incompatible licenses have been removed.

### Citation Information

If you use this dataset, please cite the original Hindawi papers as appropriate for your use case. Additionally, consider citing this dataset:

```
@dataset{hindawi_academic_papers_filtered,
  title={Hindawi Academic Papers Dataset (CC BY 4.0 Compatible)},
  author={Mariusz Kurman},
  year={2025},
  publisher={Hugging Face},
  note={Filtered compilation from Hindawi Publishing Corporation papers - CC BY 4.0 compatible subset}
}
```

### Contributions

Dataset compiled and structured from publicly available Hindawi academic papers. Original content copyright belongs to the respective authors and Hindawi Publishing Corporation. This dataset has been filtered to ensure CC BY 4.0 compatibility.

### Dataset Card Creation

This dataset card was generated on 2025-06-09 and contains statistics based on the filtered dataset of 299,316 papers (filtered from 299,363 original papers).