Inspiration
Various private service chatbots
What it does
Answer questions about registering with a fictional University
How we built it
Python. nltk and tensorfow.keras for model. Django for GUI.
Challenges we ran into
Everything was pretty straightforward for the ML model. Django was where the trouble was, mostly frontend and where to put resources. Am not too experienced with web design.
Accomplishments that we're proud of
Actually built a chatbot, even if it's just a proof of concept. Think it's very scalable. It trains very fast, all it needs is a database of inputs and responses. Not exactly chatGPA but is a nice starting point.
What we learned
Never really worked with natural language processors before. Learned a bit about tokenization and bag of words models.
What's next for stockUniversityChatBot
Scale it up. Add more functionalities. Maybe add some Twilio automatic voice calls?
Log in or sign up for Devpost to join the conversation.