competition
complete
$5,000

Woohoo! This competition has come to a close!

Many thanks to the participants for all of their hard work and commitment to using data for good!

Wild bees are important pollinators and the spread of colony collapse disorder has only made their role more critical. Instead of experts needing to go through countless bee photos by hand, the winning algorithm can help automate this process and save a huge amount of valuable time.

Why

Gaining insight into changing bee populations is critical to understand the effects of colony collapses. Unfortunately, it takes a lot of time and effort for researchers to gather data on wild bees. Although BeeSpotter is making this process easier by using images submitted by citizen scientists, they still require experts to go through each image to examine and identify each bee. This obviously is very time consuming.

The Solution

Build an algorithm that identifies the type of bee in each photograph. METIS has asked the Drivendata community to build an algorithm that is able to succesfully identify honey bees and bumble bees in a photograph, even when those photographs have great variety in their backgrounds, positions, and image resolutions.

The Results

When we challenged our community to build an algorithm to pick out the genus of a bee based on the image, we were shocked by the results: the winners achieved a 0.99 AUC (out of 1.00) - a common measure of a model’s accuracy - on the images they had never seen before.


RESULTS ANNOUNCEMENT + MEET THE WINNERS

WINNING MODELS ON GITHUB