Although currently not using real time data, The Cabsense app by Sense Networks is an interesting project that shows potential for how data and sensor networks can be utilized in practical on the ground situations for users.
From their site:
CabSense uses data from the New York City Taxi & Limousine Commission and other sources to help New Yorkers and visitors to the Big Apple find the best nearby street corners to hail an open cab based on the day of the week, the time, and their current location.
CabSense street corner ratings are derived from machine learning algorithms applied to tens of millions of historical location points from the pickups and drop-offs of all New York City taxi cabs. Every street corner in the city is rated from 5 stars (best corners to hail an open cab) to a ½ star (low chance to hail an open cab). Corners with historically very few taxi pickups don't get any stars at all. These ratings change based on the time of day and the day of the week.
If you are in an area without yellow cab traffic, CabSense will display contact information for for-hire vehicles known as "car services" or "livery cabs". These cars are prohibited from picking up street hails, and only pick up customers who have called the car service's dispatcher and requested a car. You can click on the phone number within CabSense to call and request a pickup.
Finally, in areas of the city where there are typically few available taxis, CabSense finds and recommends nearby car services, even providing an option to call a livery cab with a single touch. Users can download CabSense for free at www.cabsense.com and the application is available from Apple’s iPhone Application Store or Google’s Android Marketplace.
CabSense was built on Sense Networks' powerful Macrosense platform that uses machine learning to transform billions of location data points into meaningful context. CabSense, soon to be open-sourced, is a unique demonstration of one of the many ways in which developers will be able to leverage the Macrosense platform.
