Via the Los Angeles Times:
The New York Police Department (NYPD) is using pattern-recognition software so analysts can compare robberies, larcenies, and thefts to hundreds of thousands of crimes logged in the department's database, finding matches faster than they would manually. The Patternizr algorithm was launched in December 2016, and NYPD assistant commissioner of data analytics Evan Levine said, "The more easily that we can identify patterns in...crimes, the more quickly we can identify and apprehend perpetrators." Levine and co-developer Alex Chohlas-Wood trained Patternizr on 10 years of patterns that the department had manually identified. Patternizr accurately reproduced old crime patterns a third of the time, and matched parts of patterns 80% of the time. The software compares factors like method of entry, type of goods stolen, and distance between crimes, and reduces possible racial bias by not counting the race of suspects when looking for patterns.