https://github.com/etienne-bobo/skimlit-nlp

The purpose of this project is to build an NLP model to make reading medical abtracts easier.
https://github.com/etienne-bobo/skimlit-nlp

keras-tensorflow multimodal-deep-learning nlp tensorflow2

Last synced: 3 months ago
JSON representation

The purpose of this project is to build an NLP model to make reading medical abtracts easier.

• Host: GitHub
• URL: https://github.com/etienne-bobo/skimlit-nlp
• Owner: Etienne-bobo
• Created: 2023-08-20T15:43:36.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-08-20T16:07:39.000Z (over 1 year ago)
• Last Synced: 2025-01-10T10:30:01.673Z (4 months ago)
• Topics: keras-tensorflow, multimodal-deep-learning, nlp, tensorflow2
• Language: Jupyter Notebook
• Homepage:
• Size: 1.95 MB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
SkimLit: Let's make [PubMed](https://pubmed.ncbi.nlm.nih.gov/) abstracts easier to read 

Pubmed is a platform where we can find millions of articles in medical fields

THe purpose of this project is to build an NLP model to make reading medical abtracts easier.

Model: Replicate the [PubMed 200k RCT: a Dataset for Sequential Sentence Classification in Medical Abstracts](https://arxiv.org/abs/1710.06071) paper

(model structure avalaible here https://arxiv.org/pdf/1612.05251.pdf)

``This is a multimodal model, and its implementation is carried out using TensorFlow-Keras.``

![Alt text](images/model.png)

Dataset: https://www.kaggle.com/datasets/matthewjansen/pubmed-200k-rtc

Input and output:

![Alt text](images/input-output.png)