A deep learning based chatbot for COVID-19
The chatbot is available at: https://huggingface.co/spaces/waleedgeo/covidbot.
The app was created using Gradio and Hugging Face.
The data for the chatbot was collected from the following sources:
- https://www.gov.uk/guidance/people-with-symptoms-of-a-respiratory-infection-including-covid-19
- https://www.nhs.uk/conditions/covid-19/covid-19-symptoms-and-what-to-do/
- https://www.fda.gov/emergency-preparedness-and-response/coronavirus-disease-2019-covid-19/covid-19-frequently-asked-questions
- https://www.coronavirus.gov.hk/
- https://www.who.int/europe/news-room/fact-sheets/item/post-covid-19-condition
- https://www.who.int/southeastasia/outbreaks-and-emergencies/covid-19/questions/post-covid-19-q-a
- https://www.nhs.uk/conditions/covid-19/long-term-effects-of-covid-19-long-covid/
- https://www.transperfect.com/dataforce/covid-19-chatbot
- https://www.ha.org.hk/visitor/ha_visitor_index.asp?Content_ID=254422&Lang=ENG&Dimension=100
The model was trained using LSTM model for 200 epochs. The model was trained using the Adam optimizer with a learning rate of 0.001. The model was trained on a GPU (NVIDIA RTX 3070, Compute Unit 8.6). The summary of model is:
The model is trained on the above mentioned sources data, and achieves an accuracy of 0.99. Furthermore, to validate the model performance, a survey will be performed. The results of the survey will be updated here.
