Machine Learning: How ML Can Detect Crime in Charlottesville; Organized Labor’s Fight Against Automation in the U.S.

Goodall, Connor, School of Engineering and Applied Science, University of Virginia
Norton, Peter, EN-Engineering and Society, University of Virginia
Vrugtman, Rosanne, EN-Comp Science Dept, University of Virginia
Morrison, Briana, EN-Comp Science Dept, University of Virginia

Companies automate to lower labor costs and increase productivity. Many workers fear
that automation threatens their jobs.

Crime prevention requires effective crime forecasting. The project team developed a
machine learning algorithm to help law enforcement agencies in Charlottesville, Virginia,
forecast violent and nonviolent crime rates. The Charlottesville open data catalog supplied data
for a training set and a test set. We used eight regression models for the training set, tuned their
hyperparameters, and chose the one with the lowest error rate as the model for the algorithm. The
new algorithm gave 0.016% and 0.85% as the distances between predicted and actual violent and
nonviolent crime rates. The algorithm may be useful in a visualization tool predicting local crime

Many workers fear automation threatens their jobs. In the United States, unions help their
members resist or manage the employment threats of automation by publicizing the problem,
through influencing the government, through picketing and boycotting, and by helping their
members adapt through training or reskilling programs or by researching ways how to adapt.

BS (Bachelor of Science)
Automation, Labor Unions, United States, Machine Learning, Crime Rates, Crime, APA

School of Engineering and Applied Science

Bachelor of Science in Computer Science

Technical Advisor: Briana Morrison

STS Advisor: Peter Norton

Issued Date: