AI Can Now Scan News for Police Shootings Nationwide
Accurately tracking police-related fatalities has been a challenge, as various authorities and organizations maintain inconsistent data collection practices. To address this issue, researchers at the University of Massachusetts Amherst have developed an AI system that can extract information on police shootings from news reports. The system aims to create a more accurate and comprehensive database of these events, potentially shedding light on a complex and sensitive issue.
The researchers utilized a machine learning approach, training their AI to identify news articles that mention police or fatalities. They extracted relevant information from these articles, such as details about the individuals involved and the circumstances surrounding the incidents. The system was trained using the Fatal Encounters database, a manually curated collection of police-related fatalities.
While the system currently has a 57% accuracy rate in identifying police shootings, the researchers believe its performance can be significantly improved with more data and training. They emphasize that the AI is not intended to replace human oversight but rather to assist in streamlining the process of gathering crucial information.
The project has received significant attention due to its potential to shed light on a critical aspect of public safety and accountability. Its ability to extract information from news reports could provide valuable insights into the frequency and circumstances of police-related fatalities across the country.
