Datasets:

SDSC
/

open-pulse-hackathon-data-analysis

---
license: cc-by-4.0
configs:
- config_name: lauzhack-2023
  data_files:
  - split: train
    path: lauzhack-2023/*.parquet
- config_name: lauzhack-2024
  data_files:
  - split: train
    path: lauzhack-2024/*.parquet
- config_name: lauzhack-2025
  data_files:
  - split: train
    path: lauzhack-2025/*.parquet
dataset_info:
- config_name: lauzhack-2023
  features:
  - name: id
    dtype: int64
  - name: title
    dtype: large_string
  - name: description
    dtype: large_string
  - name: url
    dtype: large_string
  - name: team
    list: string
  - name: awards
    list: string
  - name: categories
    list: string
  - name: hackathon_name
    dtype: large_string
  - name: hackathon_year
    dtype: int64
  - name: hackathon_location
    dtype: large_string
  splits:
  - name: train
    num_bytes: 22974
    num_examples: 67
  download_size: 18646
  dataset_size: 22974
- config_name: lauzhack-2024
  features:
  - name: id
    dtype: int64
  - name: title
    dtype: large_string
  - name: description
    dtype: large_string
  - name: url
    dtype: large_string
  - name: team
    list: string
  - name: awards
    list: string
  - name: categories
    list: string
  - name: hackathon_name
    dtype: large_string
  - name: hackathon_year
    dtype: int64
  - name: hackathon_location
    dtype: large_string
  splits:
  - name: train
    num_bytes: 28812
    num_examples: 79
  download_size: 21747
  dataset_size: 28812
- config_name: lauzhack-2025
  features:
  - name: id
    dtype: int64
  - name: title
    dtype: large_string
  - name: description
    dtype: large_string
  - name: url
    dtype: large_string
  - name: team
    list: string
  - name: awards
    list: string
  - name: categories
    list: string
  - name: hackathon_name
    dtype: large_string
  - name: hackathon_year
    dtype: int64
  - name: hackathon_location
    dtype: large_string
  splits:
  - name: train
    num_bytes: 39525
    num_examples: 67
  download_size: 28942
  dataset_size: 39525

tags:
- hackathon
- projects
- sdsc
---

# LauzHack Projects Dataset

## Dataset Summary

This dataset contains comprehensive information about projects submitted to
**LauzHack** (EPFL's student-run hackathon) from 2023 to 2025. Each project
includes details about the project title, description, team members, awards, and
categories.

LauzHack is an annual 24-hour hackathon hosted at EPFL (École Polytechnique
Fédérale de Lausanne) in Lausanne, Switzerland, bringing together students and
hackers to create innovative solutions within a short timeframe.

## Dataset Details

- **License:** CC-BY 4.0 (Creative Commons Attribution)
- **Language:** English
- **Data Source:** LauzHack official website and event submissions
- **Last Updated:** February 2026

## Dataset Structure

### Data Instances

Each instance represents a single hackathon project submission with the
following fields:

| Field                | Type         | Description                                               |
| -------------------- | ------------ | --------------------------------------------------------- |
| `id`                 | int64        | Unique identifier for the project                         |
| `title`              | string       | Project title/name                                        |
| `description`        | string       | Project description and overview                          |
| `url`                | string       | Link to the project (Devpost, GitHub, or project website) |
| `team`               | list[string] | List of team member names                                 |
| `awards`             | list[string] | Awards/prizes won by the project                          |
| `categories`         | list[string] | Project categories (e.g., "Best AI", "Most Creative")     |
| `hackathon_name`     | string       | Name of the hackathon event                               |
| `hackathon_year`     | int64        | Year the hackathon took place                             |
| `hackathon_location` | string       | Location of the hackathon                                 |

### Data Splits

The dataset is available in multiple configurations (subsets) for different
years:

- **lauzhack-2023**: 67 projects from LauzHack 2023
- **lauzhack-2024**: 79 projects from LauzHack 2024
- **lauzhack-2025**: 67 projects from LauzHack 2025

## Usage

### Loading the Dataset

```python
from datasets import load_dataset

# Load a specific year's data
dataset = load_dataset("username/lauzhack-projects", "lauzhack-2025")

# Access the training split
projects = dataset["train"]

# View a sample project
print(projects[0])
```

### Example Project

```python
{
    "id": 1,
    "title": "AI-Powered Study Assistant",
    "description": "An AI chatbot designed to help students with course material...",
    "url": "https://devpost.com/software/...",
    "team": ["Alice Dupont", "Bob Mueller"],
    "awards": ["Best Use of AI"],
    "categories": ["Machine Learning", "Education Tech", "Open Science"],
    "hackathon_name": "LauzHack 2025",
    "hackathon_year": 2025,
    "hackathon_location": "EPFL, Lausanne, Switzerland"
}
```

## Dataset Creation

### Data Collection

Projects were extracted from the LauzHack official website using web scraping
techniques. The data includes all publicly submitted projects from past events.

### Data Processing

- Projects are cleaned and deduplicated
- Team members and awards are extracted into list structures
- URLs are validated and normalized
- Text fields are preserved in their original language

## Considerations for Using the Data

### Intended Use

- **Academic Research**: Student hackathon trends, innovation patterns
- **Career Analysis**: Skills and technologies used in recent projects
- **Event Organization**: Understanding project scope and diversity
- **Machine Learning**: Training models for project classification and
  recommendation

### Known Limitations

- Only includes publicly available project data
- Some projects may have incomplete information (missing descriptions, URLs, or
  award data)
- Team member names as submitted (may include pseudonyms)
- Hackathon location fixed as EPFL; some projects may have had distributed teams

### Ethical Considerations

- All data is from public hackathon submissions
- Licensed under CC-BY 4.0 (attribution required)
- No personal information beyond team member names included
- Respects original project creators' intellectual property

## Citation

If you use this dataset, please cite:

```bibtex
@dataset{lauzhack_projects_2026,
  author = {Tir Raazia, Eisha},
  title = {Hackalysis Dataset 2023-2025},
  year = {2026},
  license = {CC-BY-4.0},
  url = {https://huggingface.co/datasets/username/lauzhack-projects}
}
```

## Additional Information

### Related Resources

- [LauzHack Official Website](https://lauzhack.com/)
- [EPFL Innovation Park](https://www.epfl.ch/)
- [hackathon-analysis GitHub Repository](https://github.com/sdsc-ordes/open-pulse-hackalysis)

### Dataset Maintenance

This dataset is maintained by the open-pulse-hackalysis project

---