EC-602-Twitter-API-Project
Problem Statement?
Patients POV:
- Long Waiting times.
- Unavailability of medical staff.
- Heightened Anxiety and Nervousness.
Medical staff’s POV:
- Prioritizement of medical emergencies.
- Poor Management.
Project mission?
I want to gage the user expereince in hospitals.
Users?
Patients POV:
- Too many unanswered queries due to epidemics surges such as Covid etc.
- Long Waiting times due to unavailability of medical staff and increased traffic in hospital.
- Unavailability of medical staff.
- Heightened Anxiety and Nervousness in patients due to long waiting times, expectation of results leading to patients becoming paranoid.
Performing sentiment analysis on the data gathered from responses from patients to visualize the emotions into positive, negative and neutral categories.
List user stories?
I want the patients experiences that are sent by patienst through tweets be gaged though sentiment analsysis.
MVP (minimum Valued Product):
- Get Method which fetches all the data from the twitter related to certain kewords
- Sentiment analaysis method which performs sentiment analsysis on data and shows it through visualizations
Tools utillized
- Twitter API
- Vader for performing sentiment analysis
- sns scrape library to scrape last tweets from the twitter api easily
Future Scope:
-
Incorporate sentiment analysis and construct a chatbot to respond to patient enquiries based on the sentiment of the preceding remark or query.
-
Adding more databases of diseases and its related attributes to expand the dataset used by the chatbot will help make the chatbot more generalizable in a clinical setting.
-
Implementing a dashboard
-
improved NLU to make it more robust.
-
Incorporate sentiment analysis into the chatbot to respond to patient enquiries based on the sentiment of the preceding remark or query.
-
Adding more databases of diseases and its related attributes to expand the dataset used by the chatbot will help make the chatbot more generalizable in a clinical setting.
-
Improving the dashboard by implementing Age based filters using actions that will change the attributes of the health conditions dynamically according to the age filter selected and will showcase some and hide other features at the time of selection.