Sentiment was a Hackathon project made at Hack The 6ix. The project's goal was to tracks a meeting conversation and find its sentiment. Using the transcripts of the meeting and an open source linguistic analysis tool, we were able to successfully build the project in React and Node.js. After 36 hours of coding, we gave our presentation to the judges and won first place.
Aug 2016 - Dec 2016
AI Chatbots have been around for a while but have been traditionally based on keywords. This project aimed to have the chatbot understand the sematic contents of the messages by processing the messages with a neural network. The neural network was a Seq2Seq model, the state of the art at the time, and was moderately successful at processing chat messages.