Articles by John Kristian
This is a part of our continuing series on engineering and analytics at LinkedIn. If this isn’t your cup of Java, check back tomorrow for regular LinkedIn programming. Else, check out our Engineering Blog for more. - Ed.
There are over 1 million groups on LinkedIn and millions of members join these groups every week. When working with this volume of data, we ran into an interesting problem: how can we rank group discussions so that the most “interesting” ones appear at the top? What algorithm could handle the huge stream of activity – new discussions, comments, likes, views, follows and shares – to produce an order that would work for active and inactive groups as well as for active and inactive members?