This Google-Assistant integrated application allows the Women users to speak with the Agent about their problems. Problems that could not be shared with friends or family but the assistant.
What it does
The user speaks to the assistant regarding her problem. This information is used to find out the emotion behind the conversation. It calculates a score that is further combined with the Persona. Persona is an analysis that our app aims to perform on our users. We know that not everyone appreciates similar kind of replies. For some may want to just talk about their problem, on the other hand someone would just want a friend who could talk to them. For each user, there exists a short survey that allows the assistant to understand what kind of responses are expected. The combined score is then integrated with the dataset that has been filtered/trained such that agent only gives out positive affirmation.
How we built it
We divided our work into three major parts: 1) Webapp to register a user with their profiles 2) NLP sentiment analysis using dialogflow and GCP 3) API to generate positive affirmations for a particular user
Challenges we ran into
Connecting dialogflow with Google Cloud Platform Deploying flask server to Heroku NLP using Google Cloud Platform Getting credentials for login and sign up Integrating all the components
Accomplishments that we're proud of
We learned so many new technologies that we didn't even know before just to get this project done.
What we learned
React, NodeJS, Dialogflow and fulfillments, webhooks, Google Cloud Platform and its NLP API, Heroku, Python Flask, Firebase, Oauth0 API creation,integration,updation
Creating a better and smoother user experience to deliver more personalized features with sign in capabilities