Ambience
NLP | Chat | Text classification
Problem:
Chats with friends are difficult. They contain misunderstandings, a constant back and forth of ideas, suggestions, questions, explanations, etc. Is there some way to get a high-level overview of what is actually happening in these chats?
Solution
Ambience is a WhatsApp Chat Analysis WebApp. Given you have a long English chat conversation, you can export and have it analyzed. For each single message in the chat, ambience predicts an intent and provides an overview of used intents. This can provide an overall impression which chat participant is more likely to propose something, or who asks more questions.
My Role
I built a messenge classifier that is trained on few samples of chat history and forum chats. Because of the lack of training data, the performance of the message classification isn't stellar. In this sense, this project is a proof of concept how an AI-based chat analysis tool would look like. Find the code on GitHub and play with the live version.
Technologies:
Rasa NLU, Python, React, Flask
