https://github.com/shreenidhi7700/skimlit-project

Natural Language Processing Project -SKIMLIT
https://github.com/shreenidhi7700/skimlit-project

google-colab lstm-neural-networks numpy pandas python3 rnn-tensorflow tensorflow

Last synced: 3 months ago
JSON representation

Natural Language Processing Project -SKIMLIT

• Host: GitHub
• URL: https://github.com/shreenidhi7700/skimlit-project
• Owner: shreenidhi7700
• Created: 2025-01-18T14:54:31.000Z (over 1 year ago)
• Default Branch: master
• Last Pushed: 2025-01-19T06:36:20.000Z (over 1 year ago)
• Last Synced: 2025-03-14T02:11:54.953Z (over 1 year ago)
• Topics: google-colab, lstm-neural-networks, numpy, pandas, python3, rnn-tensorflow, tensorflow
• Language: Jupyter Notebook
• Homepage:
• Size: 70.8 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# SKIMLIT Project

SkimLit is a natural language processing (NLP) project aimed at making the reading of medical abstracts more accessible. This project replicates the methodology outlined in the paper "PubMed 200K RCT: a Dataset for Sequenctial Sentence Classification in Medical Abstracts," using TensorFlow and various deep learning techniques.

The primary aim of SKIMLIT is to classify different parts of academic publications' abstracts into specific categories. This can be particularly useful for researchers, students, or anyone interested in quickly understanding the key components of a scientific paper without having to read the entire document.

Dataset is available over here: https://github.com/Franck-Dernoncourt/pubmed-rct

The dataset consists of approximately 200,000 abstracts of randomized controlled trials, totaling 2.3 million sentences. Each sentence of each abstract is labeled with their role in the abstract using one of the following classes: background, objective, method, result, or conclusion.