We would like to build a chatbot that is able to recommend a suitable cocktail for users who are not sure what they want. After a few questions, the chatbot will then return a cocktail recipe that is chosen base on the users' input.
- Collect : Plain text from online cocktail recipes
- Prepare:
- Phase 1 - Encode the text data to XML file and XML schema
- Phase 2 - Import the XML into a database
- Access: A chatbot that asks users’ preferences ( taste, base, hashtags, juice, syrup) and access the database with SQL queries created based on users’ input.
- cocktails.xml: xml file for cocktail recipes
- xml_schema.xml: xml schema for validating cocktails.xml
- XMLtoDB.py: The script for extracting and inserting xml data into database
- cocktail_db.sql: The database of cocktail recipes. Tables include:
- cocktail(name, hashtags)
- flavor(name, taste1, taste2)
- flavor_minor(name, taste3, taste4)
- ingredients(name, alcohol, additional)
- cocktail_chatbot.py: Python code for the chatbot using loops and SQL queries
- cocktail_chatbot_ver2.py: Python code that is similar to cocktail_chatbot.py, but with a function that creates SQL queries in a more dynamic fashion
- Cocktail_Presentation Slides: Slides used for group presentation