Inspiration
Our inspiration was our AI/LI subjects that made us use rule-based ai, as well as learning front end.
What it does
The users have to answer a questionnarie where the values are on a prefixed range of their preferences. After their data is collected, the rule-based ai generates a list of possible destinations.
How we built it
We started by designing a pipeline with some data recolection, filtering, more data through the API and final grading. For this we made use of Prolog for the logic, python for scripting and grading, javascript and node for frontend and a couple other things.
Challenges we ran into
We knew little of prolog, nothing of APIs and frontend. 80% of the code wouldn't be understandable for us last week.
Accomplishments that we're proud of
We learned enough to write code from which 80% we wouldn't understand a week ago.
What we learned
Overall project planning and structuring, streamlining data through JSON and CSV files, API handling in Python, some frontend and advanced Prolog.
What's next for SkyScanner the perfect reunion. FeatProlog ¡TOMAMANGO!
A lot of improvements could be made: we failed to build a loading screen, users could select more than 1 desired destination, we could have built a database instead of hardcoding thousands of rules in Prolog...
Log in or sign up for Devpost to join the conversation.