Added: July 4, 2015 – Last updated: July 4, 2015


Authors: Ellie Clougherty, John Clougherty, Xiaoqian Liu, and Donald Brown

Title: Spatial and Temporal Analysis of Sex Crimes in Charlottesville, Virginia

Subtitle: -

In: 2015 IEEE Systems and Information Engineering Design Symposium

Edited by:

Place: Charlottesville, VA

Publisher: IEEE

Year: 2015

Pages: 69-74

ISBN-13: 9781479918317 – Find a Library: Wikipedia, WorldCat

Language: English

Keywords: 20th Century, 21st Century | U.S. History


Link: EDAS (Free Access)

Link: IEEE Xplore Digital Library (Restricted Access)


Authors: Donald E. Brown, Data Science Institute, University of Virginia

Abstract: »Sexual assault and interpersonal violence affects university communities in disproportionate numbers to those of the general population. It is estimated that one in five women will be the victims of a sexual assault during their college years. In this study, we use kernel density estimation, logistic regression and random forest modeling to conduct spatial and temporal analysis of sexual assault at the University of Virginia and wider Charlottesville community between 1990 and 2015. This paper takes into account the cultural characteristics of the community to explore the underlying reasons for concentrations of criminal sexual assault behavior. Our results show that proximity to registered sex offenders and Greek life residences are the two most important predictors of sexual assault crime among all the variables studied. These spatial and temporal models provide insights to better understand crime patterns and improve policing and response.« (Source: IEEE Xplore Digital Library)


  Abstract (p. 69)
  Introduction (p. 69)
  Literature Review (p. 69)
    I. Sexual Assault (p. 69)
    II. Spatial and Temporal Models (p. 70)
  Data (p. 70)
    I. Sexual Assault Incident Data (p. 70)
    II. Spatial Variables (p. 70)
    III. Sex Offender Registry Data (p. 71)
    IV. Weather Data (p. 71)
  Methodology (p. 71)
  Results (p. 71)
    I. Kernel Density Estimation Maps (p. 71)
    II. Random Forest (p. 72)
    III. Logistic Regression (p. 72)
    IV. Time Series (p. 73)
  Conclusion (p. 73)
  Future Work (p. 74)
  Acknowledgment (p. 74)
  References (p. 74)
  Author Information (p. 74)