New Yorker applied machine learning to blocked bike lane problem

It took Bell around three weeks to develop the algorithm and his system took about a day to analyze the traffic cam footage. The results showed that the bus lane on the block covered by the camera was blocked 57 percent of the time while the two bike lanes were blocked 40 percent of the time. Based on those numbers, Bell determined that approximately 850 vehicles had blocked the bike lane during those 10 days and 1,000 blocked the bus lane. And while these findings are based on just one city block and a short period of time, the other 101 miles of bus lanes and 435 miles of bike lanes throughout the city suffer from the same issues. It's a problem many people want a solution to.