SEATTLE -- Seattle police want to speak with a University of Washington student, but not because he's breaking the law.
Jay Feng, an electrical engineering major, published a story on his blog plotting and predicting crime in the city. Feng used public data that he accessed on the city's website.
Feng learned the Rainier Beach Neighborhood and Aurora Avenue, north of Green Lake, had the most dangerous crimes over the past several years. He also found out that the Sand Point and Lake City neighborhood had the most false alarm 911 calls.
"The ones that have the most false alarms are within the bottom ten in how much crime they get," Feng said. "They're pretty crime-free neighborhoods."
Feng knows police already have crime analysts that investigate this type of situation, but he hopes that the department considers this when they determine what calls they respond to first.
He knows all 911 calls have to be taken seriously. But Feng believes that if police use the data that is out there, they can respond to calls in a logical order during peak times.
"Obviously, I think if there's like a fire, an emergency, obviously they'll go right away But possibly in neighborhoods with more false alarms, they might let it relax a little because they know what they are going to isn't as urgent," Feng said. "My idea of it was, if there was a good enough crime prediction software where you can analyze how old the person is, tone of voice, and where they are located, you can feed into an algorithm and spit back to the police officer or be like, 'be cautious', or 'this has a high chance of being a false alarm.'"
Feng also found that the number of police responses dropped during lunch and dinner break. He wonders if that is because officers don't get out faster during meal or shift breaks or if there are just less crimes that happen during those hours.
The Seattle Police Department wants to talk to Feng as they continue to make a push at better transparency. SPD has said for months now that they want a number of tech minds to come forward if they believe they can help the department improve how they make information available to the public.
If you are interested in learning more about Feng's findings you can find them on his blog.