Inspiration
We first attempted to build a recommender system using GCP Vertex AI but faced deployment challenges. This led us to pivot toward building a deep learning chatbot that could reason over stories.
What it does
The chatbot reads a story, takes a user’s question, and responds whether the question is correct or relevant to the story. All interactions follow a Q&A format.
How we built it
I used Python notebook with TensorFlow, Keras, Matplotlib for visualization, numpy and pickle.
Challenges we ran into
Vertex AI integration was difficult for the initial recommender project. Creating effective training data and achieving accurate contextual understanding also took effort.
Accomplishments that we're proud of
We successfully built a context-aware chatbot that can interpret story-based information and classify questions accurately. Pivoting the project was a key achievement.
What we learned
We learned the importance of flexible project scoping and how deep learning can outperform rule-based systems in language understanding. Different types of chatbots out there.
What's next for Chatbot-with-DeepLearning-and-Python
Continue learning and practicing python coding and other related to data science and machine learning
Log in or sign up for Devpost to join the conversation.