Daily Papers Analysis

This project aims to analyse the top papers in the AI/ML space. We will use the listed papers in Hugging Face Daily Papers for the past year and study the relationship between the papers, authors, and their organizations.

Datasets

All the details of our target papers is saved as Assets\papers_details.json. These details are used to create the datasets we need to construct the graphs. Our graph datasets are in GraphDataset folder:

• paper.csv: Contains paper information including ID, publication date, and title
• author.csv: Contains author information including ID and name
• paper_author.csv: Maps papers to their authors (many-to-many relationship)
• paper_category.csv: Maps papers to their research categories (many-to-many relationship)
• category.csv: Contains category information including ID and category name/field