Deep Paper Predictor

Welcome to the Deep Paper Predictor GitHub repository! This repository contains code for a web application that assists researchers in finding relevant research papers and predicting the subject area of a paper based on its abstract.

Overview

The Deep Paper Predictor is a web application developed using Streamlit, TensorFlow, PyTorch, and Sentence Transformers. It consists of two main functionalities:

• Paper Recommendation: Given the title of a research paper, the application recommends similar papers based on cosine similarity scores calculated using sentence embeddings.
• Subject Area Prediction: Given the abstract of a research paper, the application predicts the subject area using a shallow Multi-Layer Perceptron (MLP) model.

How to Use

To use the Deep Paper Predictor:

1. Clone this repository to your local machine.
2. Install the required dependencies listed in the requirements.txt file.
3. Need to save all the models form the Research Paper recommendation and subject area prediction using sentence transformer.ipyb file.
4. There will be total 7 files in the models folder so be sure that you have save all the 7 models.
5. Run the Streamlit application by executing streamlit run app.py in your terminal.
6. Enter the details of the paper (title and abstract) in the sidebar.
7. Click on the "Recommend" button to get recommendations and predicted subject areas.

Dependencies

• TensorFlow
• PyTorch
• Sentence Transformers
• Streamlit
• Googlesearch

All dependencies can be installed via pip using the command pip install -r requirements.txt.

File Structure

• app.py: Main Streamlit application file containing the user interface and interaction logic.
• models/: Directory containing saved models, embeddings, and configurations.
• utils.py: Utility functions for recommendation and subject area prediction.

Contributors

• Devanshu