If that's all it does, then it's not very impressive. You can trivially create a program that finds nearby Chinese restaurants with open parking and WiFi that's kid friendly, by creating a table of all such establishments and their "nearby" locations.

Presumably, this was supposed to prove something else- what else? How did they measure that "90% success rate"?

Just throwing percentiles in the air like that says nothing and makes the whole thing appear not a little dodgy.