Palaround is a YC startup that helps user in making new friends. In their words: “Palaround is a platonic friending app engineering compatibility and serendipity. It’s all about finding the perfect friends for you that you didn’t even know existed. Because who knows, your best friend may live right next door”.
They are about to release a new version of the app called With, specially designed for event organizers and group admins. Along with this exciting new feature, they need to use Data Science to increase their matching (link prediction) algorithm.
Link prediction is a well studied problem in social network analysis with plenty of research and frameworks for it. However, the secret sauce of major social networks algorithms is, of course, a private key asset, which is an example of the value of Data Science. At 7Puentes we are aware of that and designed a novel and custom blend of techniques to understand Palaround users and their needs.
We’ve built our prototypes in popular networkx framework and hopefully will be migrating to Spark once we’ve reached our critical capacity.
With this new algorithm Palaround will be able to see the network grow organically and efficiently.