Inspiration
The main goal of this project was to learn more about how to leverage APIs like Messenger API and Azure, as well as spread the good word of ABBA. Special thanks to Sarah Colquhoun for providing lyrics data and inspiring me with her ABBA Gold Twitter bot.
What it does
ABBA Gold Bot takes in a user's message, runs sentiment analysis and keyword search on the sentence via the Azure Cognitive Services resource, then returns a snippet of lyrics based on the data collected, looking at attributes like type of question, overall sentiment, presence of certain keywords, etc.
How I built it
I used the Messenger API to host the chatbot, alongside Flask and ngrok to keep a server running. Azure was used for NLP like sentiment analysis.
Challenges I ran into
It was my first time ever using Azure or the Messenger API, so understanding how to properly set up these APIs before even a single line of code was written was quite a headache.
Accomplishments that I'm proud of
-I know more about the Messenger API, Azure, and building chatbots in general now! -While cryptic at times, the chatbot works smoothly and responds to user queries in a timely manner -I got to raise awareness about the best band of the 20th century
What I learned
-While requiring some learning for first-time users, services like Azure can save a lot of time for implementing stuff with relatively little overhead cost -Word analysis is a lot harder than it seems at first, even with backup from things like Azure
What's next for ABBA Gold Chatbot
-Adding more ABBA songs and relevant keyword tags -Using better NLP techniques to further automate the lyric selection process
Log in or sign up for Devpost to join the conversation.