US Fatal Police Shooting Analysis and Prediction: A Study

Authors: (1) Yuan Wang, University of Rochester (e-mail: yuan.wang1@simon.rochester.edu); (2) Yangxin Fan, University of Rochester (e-mail: yfan24@ur.rochester.edu). Table of Links Abstract and Introduction Related work Methodology Media reporting analysis WP fatal police shooting dataset insight Fatal police shooting rate and victims race prediction Conclusion and References Abstract We believe that “all men are created equal” [11]. With the rise of the police shootings reported by media, more people in the U.S. think that police use excessive force during law enforcement, especially to a specific group of people. We want to add our two cents point of view by multidimensional analysis to reveal more facts than the monotone mainstream media. The more facets we understand the problem, the better solution that the whole society could approach. 1. Introduction Our report has three parts: First, we analyzed and quantified fatal police shooting news reporting deviation of mainstream media. Second, we used FP-growth to mine frequent patterns, clustered hotspots of fatal police shootings, and brought multi-attributes (social economics, demographics, political tendency, education, gun ownership rate, police training hours, etc.) to reveal connections under the iceberg. Third, we built regression models based on correlation analysis for numeric variables selection to predict police shooting rates at the state level. We also built classification models based on Chi-square testing for categorical variables selection to predict the victims’ race of fatal police shootings. The main datasets we choose for our analysis include: 1. Washington Post Fatal Police Shooting Dataset (WP data) [13] covers fatal police shooting from 01-01-2015 to 12- 02-2020. 2. KilledByPolice (KBP): Fatal police shooting reported in KilledByPolice website [6] from 01-01-2015 to 11-04-2020. This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. Authors: (1) Yuan Wang, University of Rochester (e-mail: yuan.wang1@simon.rochester.edu); (2) Yangxin Fan, University of Rochester (e-mail: yfan24@ur.rochester.edu). Authors: Authors: (1) Yuan Wang, University of Rochester (e-mail: yuan.wang1@simon.rochester.edu); (2) Yangxin Fan, University of Rochester (e-mail: yfan24@ur.rochester.edu). Table of Links Abstract and Introduction Abstract and Introduction Related work Related work Methodology Methodology Media reporting analysis Media reporting analysis WP fatal police shooting dataset insight WP fatal police shooting dataset insight Fatal police shooting rate and victims race prediction Fatal police shooting rate and victims race prediction Conclusion and References Conclusion and References Abstract We believe that “all men are created equal” [11]. With the rise of the police shootings reported by media, more people in the U.S. think that police use excessive force during law enforcement, especially to a specific group of people. We want to add our two cents point of view by multidimensional analysis to reveal more facts than the monotone mainstream media. The more facets we understand the problem, the better solution that the whole society could approach. We believe that “all men are created equal” [11]. With the rise of the police shootings reported by media, more people in the U.S. think that police use excessive force during law enforcement, especially to a specific group of people. We want to add our two cents point of view by multidimensional analysis to reveal more facts than the monotone mainstream media. The more facets we understand the problem, the better solution that the whole society could approach. 1. Introduction Our report has three parts: First, we analyzed and quantified fatal police shooting news reporting deviation of mainstream media. Second, we used FP-growth to mine frequent patterns, clustered hotspots of fatal police shootings, and brought multi-attributes (social economics, demographics, political tendency, education, gun ownership rate, police training hours, etc.) to reveal connections under the iceberg. Third, we built regression models based on correlation analysis for numeric variables selection to predict police shooting rates at the state level. We also built classification models based on Chi-square testing for categorical variables selection to predict the victims’ race of fatal police shootings. The main datasets we choose for our analysis include: 1. Washington Post Fatal Police Shooting Dataset (WP data) [13] covers fatal police shooting from 01-01-2015 to 12- 02-2020. 2. KilledByPolice (KBP): Fatal police shooting reported in KilledByPolice website [6] from 01-01-2015 to 11-04-2020. This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. available on arxiv