Skip to content

An advanced web application specialized in the automatic classification of medical abstracts. Leveraging natural language processing (NLP) techniques.

Notifications You must be signed in to change notification settings

Abdelhamid2c/ClaMeD-Classification-WebApp

 
 

Repository files navigation

A web application for the classification of medical abstracts


screen Homepage



Description:

This repository houses the source code of an advanced web application designed for the automatic classification of medical abstracts. The application leverages natural language processing (NLP) techniques and machine learning modeling to precisely analyze and categorize medical article summaries.

Key Features:

  • Automatic Classification : Utilizes machine learning models to classify medical abstracts into specific categories.
  • Intuitive User Interface : User-friendly web interface allowing users to submit abstracts, visualize classification results, and obtain relevant insights.
  • Section Identification : Development of specific algorithms to identify key sections of a document, such as objectives, methods, results, conclusions, etc.

screen App

Technologies Used:

  • Natural language processing (NLP) : Long Short Term Memory (LSTM)
  • Machine Learning Framework : TensorFlow

Benefits:

  • Time Savings : Researchers can save time by quickly identifying relevant sections of a document.
  • Machine Facilitated Navigation : The ability to navigate literature efficiently allows users to focus on the most relevant aspects of each article.
  • Research Optimization : Enhancement of scientific research efficiency by enabling more targeted and in-depth exploration.

About

An advanced web application specialized in the automatic classification of medical abstracts. Leveraging natural language processing (NLP) techniques.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 53.0%
  • CSS 19.7%
  • SCSS 18.0%
  • JavaScript 7.2%
  • HTML 1.2%
  • Python 0.9%