A question answering system for information about the Covid-19 pandemic, powered by BERT transformer models and BART text summarization.
Features User-friendly interface for submitting questions and receiving real-time predicted answers with accuracy score. Robust Q&A engine module that parses questions and utilizes deep learning models to find accurate answers Extensive knowledge base module that stores and manages a wealth of information on Covid-19
Components:
· Data pre-processing
· Exploratory data analysis
· Knowledge base in Elasticsearch
· BERT QA Engine
. BART text summerizer
· Frontend with streamlit
Prerequisites Python 3.6 or higher Hugging Face transformers library
Aknowledment: This project was built and updated based on the Udemy course 'Deep Learning NLP: Build and Deploy a BERT COVID Q&A System' by Deeplearnai team.