sleeping-ai/Wattpad-Complete-data

Wattpad Complete Data - Research Dataset

The Wattpad Complete Data is a comprehensive research dataset that includes all the stories published across the 18 categories available on Wattpad. While this dataset comprises 20,169 stories, it is difficult to ascertain whether it represents a fully exhaustive collection of all stories on Wattpad. This dataset was curated by identifying the stories that are accessible to a standard user, meaning that these 20,169 stories are the only ones visible to an average user.

The dataset was compiled in November 2024, and it should be noted that additional stories may have been added to the platform after the compilation date.

Purpose of the Dataset

Wattpad stands as one of the largest repositories of authentic, original, and imaginative romantic stories, with over 93 million active users monthly. The platform offers unique narratives that differentiate it from mainstream literature, thus catering to a niche genre of readers who seek distinctive storytelling experiences.

In the fields of Machine Learning (ML) and Natural Language Processing (NLP), this dataset holds immense potential. It is valuable for training domain-specific models, developing specialized language models, and conducting research on cultural, ethical, and social aspects inherent in the stories.

This dataset represents one of the first state-of-the-art (SOTA) releases from Sleeping AI.

Terms of Use

The dataset is available for non-commercial research purposes only. Users are prohibited from using the dataset to develop products that may directly compete with Wattpad or infringe on the copyright of individual story authors.

Restrictions:

• The dataset may not be redistributed, shared, or used to create personal or institutional copies.
• Upon completion of research, all copies of the dataset must be permanently deleted from repositories.
• Researchers are required to submit a summary report of their use case and published research findings. We acknowledge potential concerns regarding competitive advantages; therefore, sharing a high-level overview of the research is deemed sufficient.

By accessing this dataset, you agree that Sleeping AI and affiliated entities do not assume responsibility for any misuse by the research institution or individual researcher. While we understand that copyright laws may vary across jurisdictions, we still permit access to this dataset regardless of these differences.

How to Access the Dataset

Access is currently being granted selectively to research institutions and trusted collaborators due to the sensitive licensing nature of the dataset. To request access, please adhere to the following procedure:

1. Review the usage guidelines and restrictions outlined in this document.
2. Complete the Google Form to formally request access.
3. If urgent access is required, please contact us directly via email at [email protected].

We may request additional information before granting access. Institutional email addresses must be used when submitting access requests.

Available Data

The dataset includes the following data for each story:

• Cover images of the stories
• HTML files of all chapters
• Converted markdown of the HTML pages
• Each story is organized into its own folder

Acknowledgments

We would like to express our sincere gratitude to our collaborators, especially Fan Xiaoya from Xinyang Normal University, for their invaluable support in this endeavor.

Legal Disclaimer

This dataset is being released strictly for non-commercial and research use. Before using the dataset, please review the relevant copyright laws and statutes:

• European Union Copyright Guidelines

The dataset is released under European law and complies with the applicable legal framework.

---

By accessing and using this dataset, you acknowledge and agree to comply with the terms and conditions outlined herein.