I've been playing Pokemon Go since it released, and I must say that I've grown tired of catching Pidgeys and Rattatas. However, finding rarer Pokemon can be time-, battery-, and data-consuming. The social aspect of the game has helped me at least find some rarer Pokemon, even if I was unable to catch them for one reason or another.
In fact, just a few nights ago I spotted another player at a Pokestop with me, and we exchanged recent sightings of a Golduck and a Nidoqueen. It was awesome to have someone let me know that I could go a little ways up the street and grab something I had not caught yet. However, if he hadn't told me about it, I would have likely gone home without thinking to stop there and search around. It would be great if there was a way to locate and signify the locations of rarer Pokemon to cut down on time spent looking for them.
Enter IBM Watson and Michael Hsu. Shortly after the app was released, Hsu competed in the AT&T Shape Tech Expo Hackathon in San Fransico and decided to use the IBM Watson API to find and share the location of rarer Pokemon. Hsu was able to get Watson to locate Pokemon in the app by taking screenshots intermittently. Then, Watson would identify any Pokemon nearby and notify the player. Check it out in the video below:
Needless to say, Hsu went on to win the Best Use of Watson challenge. Maybe we'll get to see his idea manifest into a tool that we can use to help find more Magikarps down the road.