StumbleUpon Looks To Get Better At Determining When Content Is EvergreenBy: Chris Crum - August 26, 2013
StumbleUpon has an engineering contest going on, aimed at finding some help improving how it recommends content. Specifically, they’re looking for someone to develop an algorithm to make the service better at determining when a piece of content is evergreen (meaning, having relevance over a longer period of time), and when something is better off only being shown to users in a more timely fashion.
StumbleUpon has turned to Kaggle to launch the contest. Kaggle, for those unfamiliar, is a community site for data scientists. Through its Kaggle Connect platform, these scientists can compete with each other to solve data science problems put to them.
The prize for StumbleUpon’s challenge is $5,000 and a possible internship with the company. Frankly, if you can solve this problem for StumbleUpon, it seems like a safe bet they would want you as part of the team. The company did recently undergo a significant downsizing, but they’re now actively hiring.
A spokesman for StumbleUpon tells us that the service currently doesn’t consider specific categories to be evergreen or non-evergreen. As it stands, StumbleUpon simply relies on users’ ratings (thumbs up/down) to determine whether or not a piece of content is evergreen.
“Basically, any topic can have evergreen content,” he says.
In other words, it really doesn’t matter if a piece of content is submitted to the News category. It can still have long legs.
“With this contest, we hope to develop an algorithm to analyze the page itself which will let us know how evergreen it may be before we serve it to the user,” the spokesman says.
StumbleUpon says on the Kaggle page, “Many people know evergreen content when they see it, but can an algorithm make the same determination without human intuition? Your mission is to build a classifier which will evaluate a large set of URLs and label them as either evergreen or ephemeral. Can you out-class(ify) StumbleUpon? As an added incentive to the prize, a strong performance in this competition may lead to a career-launching internship at one of the best places to work in San Francisco.”
The contest began on August 16th, and will run through October 31st. After that, it remains to be seen if StumbleUpon will find its problem solved, and if so, how long it will take for the algorithm to be implemented.