Assessing the Accuracy and Ethical Implications of Automated Criminal Detection Using a Deep Learning Approach
S Jagadeesan, K. Sujigarasharma, Manjula R, T. Vetriselvi · 2023
This study examines the accuracy and ethical implications of using convolutional neural networks (CNN) for automated crime detection. A CNN model was trained on a dataset of criminal mugshots to identify potential criminal behaviour based on facial features. This study analyzed the performance of the model and achieved a high accuracy rate in identifying criminals. However, the ethical implications of automated criminal detection are also explored, including bias, privacy and human rights violations. The findings of this study highlight the need for caution and ethical considerations when implementing automated crime detection technologies. It is important to ensure that such technologies are not used to violate the rights of individuals or perpetuate societal biases.