... teaching the cops how not to alienate the people?
No, for two reasons.
First stop and frisk is based on the "broken windows" theory, which in general is sound: people take behavioral cues from what they perceive as social norms. If they look around and see people breaking windows and jumping turnstyles they'll figure everyuone does it. But stop and frisk plus being "proactive" is a case of two ideas getting together and having a bastard child. Instead of signalling what the social norms are by keeping the streets orderly, the cops are singling out individuals upon whom they will impose those norms. It's a crude exercise in behavior control, and inherenly alienating.
The second reason is Campbell's Law, roughly stated: the more you rely on a single measure to control social processes, the more that measure and the processes it controls will be corrupted. In 1995 the NYPD adopted CompStat -- a process improvement strategy based on measuring performance and holding police units like precincts accountable for their numbers. It's not a bad idea, depending on the measures chosen and if you have a critical attitude toward those measures, but the number of stops made is an inherently terrible metric. It's not a measure of success, it's a measure of activity, and is an easy number to control; if your numbers look bad you can always head out and start stopping peoiple.
There's a lot of debate over whether NYPD cops have "quotas", but it's quibbling. There doesn't have to be a quota if everyone knows bad things happen to the precinct when the numbers are low and good things happen when the numbers are high.