This is a service built for answering Frequently Answered Questions for a closed domain using Azure's Cognitive Search (read more about it here) to find answers in generally formatted documents (e.g. Product Manuals or Club Manifests). A default instance is currently deployed at our website. This service can be instantiated and customised as per need.
| Software | Version |
|---|---|
| Adobe XD | 24.0.22 |
| Zeplin | 4.0.2 |
| Python 3 | 3.7.1 |
| QnAMaker | June 2020 |
Deployed using Microsoft Azure's App Service
| Software | Tested With |
|---|---|
| Python 3 | 3.7.1 |
| QnAMaker | June 2020 |
- Clone the repository.
git clone https://github.com/MLSA-SRM/bot-gateway-rest-api
- Now install all required libraries through requirements.txt
pip install requirements.txt
- The project contains a tested live feedback service, with an option to have the feedback sent to an E-mail, or uploaded to a SQL Database (defaulted to E-mail). Code marked as
Potential Feedback Servicecan be uncommented to test the SQL-based Feedback - Now create a file with the name
.env. Add all API Keys (mentioned in following example) inside the.envfile as text. Click here to know more about hidden API Keys as Environment Variables.
SQL_USER=examplekeyvalue123
SQL_PWD=examplepassword#123
SQL_HOST=examplehost123
SQL_DB=exampledbconn#123
MAIL_USER_ID=exampleemailid
MAIL_USER_PWD=exampleemailpwd
- You can find all required keys in the
settings.py, as imported environment variables. - Lastly, update the URLs fetching the QnAMaker's API at
Line: 256inbot.js(you can find it here) as per the API credentials provided by your hosted QnAMaker Service. - Now run the Flask app
app.py
python app.py
- In your browser open http://localhost:5000 (or
:{port-number}as specified by the Flask's development server)
For more info, or having a chatbot of your own - contact us: Microsoft Learn Student Ambassadors SRM