Model Card for Model ID
Model Details
Model Description
The LongFormer Paper Classification model developed by Reem, is designed to classify research papers into four citation categories (Level 1, Level 2, Level 3, and Level 4) based on their citation counts.
Uses
The model is intended for researchers, academics, and institutions to analyze the impact and visibility of research papers based on their citation counts
Bias and Limitations
Users should be cautious of potential biases in the training data and publication practices, and validate predictions with domain expertise.
Training Details
Training Data
The model is trained on a dataset of labeled research papers with citation categories. Fine-tuning hyperparameters are used during training.
Evaluation
Evaluation metrics include accuracy, precision, recall, and F1 score across citation categories.
