Current and former Louisiana State University (LSU) computer science students have created a software company that uses data to predict the location and time of future crimes.

The company, aptly named Crimer, uses crime prediction software of the same name, which was developed by students last year as part of a computer science class project.

To develop the software, the team gathered crime data from the internet and used it as the foundation for constructing a national crime prediction map. The team applied an assortment of machine learning algorithms to the data and those algorithms extracted, transformed, loaded and predicted crime reports. Additionally, that crime data was complemented with a host of auxiliary data about weather, population, terrain and other possible contributing factors to crime.

The company is using its crime predicting software to assist law enforcement at the Lafayette Parish Sheriff’s office in Louisiana by offering crime prediction data to try to cut officer overtime and idle-time costs and increase police presence in areas likely to experience crime based on the software data. The software is also assisting the East Baton Rouge Police Department and district attorney fight the ongoing opioid epidemic with help from social media.

"We're using Twitter to find tweets about where people are soliciting opioids, then correlating those with overdose deaths," said Alexander Adams, a 2019 LSU computer science graduate and founder and chief executive officer of Crimer.

"This lets us find these damaged communities and get them the help they need through the city."

Despite the controversy of crime predicting software, which can reportedly be tainted by "dirty data," more versions of the software are either in use or in mid-development.

As such, Crimer hopes to expand its use, eventually moving into areas such as New York City and San Francisco.

To contact the author of this article, email mdonlon@globalspec.com