Having my girlfriend lives in Arizona, I constantly buying flight tickets to visit her. I read on newspaper that we should purchase around a month ahead to get the best deal, according to data. However, I still get nervous because none of the data is personalized to you, to your flight. With phly, I hope this is a good solution.
What it does
It has two REST APIs, one for web crawling and ticket number issuing, and one for spitting out the results.
It all happens in a front end website. Thanks to a HTML5 template provider, this can come true for a one-person team.
How I built it
I used Node.js and AWS to make the web scraper first. Then I started building the front end.
Challenges I ran into
Sometimes, I pursue too much about the speed of processing (usually means more code, less library used), but after finding out the cheerio npm library. It makes everything so much easier and would have saved a lot of time for a hackathon.
Accomplishments that I'm proud of
I have a machine learning interface set up and running! And I think the data is looking very good.
What I learned
how to interact and build a database.
What's next for phly
I will implement my own algorithm.