It came from the sky !
What it does
We thought how convenient would be to decide your next trips depending on your budget and especially on your mood. It would be random and fun. To be more specifically, being provided a short text in which the user is expressing his current feelings and by using NLP we are suggesting you a dream destination.
How we built it
We managed to finish most of the idea. The Stanford's NLP library was of big help, providing us with an advanced and scalable sentiment analysis algorithm.
Challenges we ran into
If I were to be honest, totally understanding and then parsing the JSON file provided by SkyScanner was more difficult than we initially expected. However, after some struggling we reached the conclusion that the data is actually very well normalised (has a good level of abstraction).
Accomplishments that we're proud of
We successfully implemented the NLP algorithm, which at the begin wasn't reliable.
What we learned
It was the first time we implemented such an advanced machine learning algorithm.
What's next for Sky Planner
Log in or sign up for Devpost to join the conversation.