Inspiration
Inpired by ImaginBank and Skyscanner challenges, we decided that creating a piggy bank that allows us to collect funds that will be destinated to travel was an interesting approach. To make the task easier, the process is done by a Bot empowered with NLP (Luis AI) that maximices the understanding capability of the ChatBot.
What it does
Flight advisor mainly allows you to:
- Manage your piggy bank with operations such as check current money, ingress money.
- Buy a flight ticket
How we built it
We have created this chatbot with microsoft bot framework for the bot itself, node.js to manage skyscanner API calls and process them to facilitate Bot's data. To make the bot "smarter" and allow the user to go faster throught the process we allow the user to be able to have a normal conversation with the bot and we'll just detect which fields are filled during the conversation process. This is done thanks to Luis.ai and the training process that we've done. To make appear images in the cards, we've done API calls to flickr to take out images related with the country we were interested in. All this application is built in a heroku.
Challenges we ran into
- Creating a bot from scratch
- NLP in a chatbot
- Dealing with Skyscanner and Flickr APIs
Accomplishments that we're proud of
Successfully delivering a chatbot in a short amount of time.
What's next for Flight Advisor
Flight advisor is just a use case of what ImagineBank can do with a chatbot, this application could be extended by adding more functionalities or simply by improving the ChatBot experience.
Built With
node.js
microsoft-bot-framework
luis.ai
heroku
javascript
skyscanner
flickr
Built With
- flickr
- heroku
- javascript
- luis.ai
- microsoft-bot-framework
- node.js
- skyscanner
Log in or sign up for Devpost to join the conversation.