The motivation behind WinnieBot was to create a scalable web-based platform where anyone can have general conversations, get any movie information, get weather updates and forecast of any global city on any specific day. In addition to that, WinnieBot is also capable of amusing the user with its jokes when he or she is feeling bored. WinnieBot uses Natural Language Processing (NLP) and Deep Learning to understand the user desires and intentions and provide him the relevant information.
- Frontend built in React/React native
- Backend part uses Node.js and Express
- Can be used on Android, iOS and web
- Understands user messages and intentions using NLP (Natural Language Processing)
- Uses REST APIs to provide relevant information to user
- Deployed on Heroku
The application/project can be run on any Android/iOS device using the Expo App. After installing the app, run the following link
https://expo.io/@onadeem/winniebot
Download the react-proj folder on your PC, install the dependencies and run it directly using react-native environment.
The working of WinnieBot is given below:
- User enters query in the message input (interface using React Native)
- WinnieBot processes the query through Deep Learning and Natural Language Processing (NLP) model using Node.js/Express
- Further Elaboration of the above step is that the query is passed through NLP model to extract important information and user intention
- And Node.js/Express backend server resolves the query and provides relevant information. Working described in the figure below
Flow control of the app is shown in figure:
The following screenshots show the working of WinnieBot:
