Tweet data analysis

advertisement
Crime analysis and policing implementation
in the space - time context
Ling Wu, Ph.D.
Computational Social Science Lab, Kent State University,
U.S. & College of Criminal Justice, Zhongnan University
of Economics & Law, China
1
About myself

Adjunct professor, Computational Social Science Lab, Kent State University,
U.S.

Associate professor, College of Criminal Justice, Zhongnan University of
Economics & Law, China

Ph.D. Criminal Justice, Sam Houston State University, U.S. (2012)

LL.M. International & Comparative Law, Uppsala University, Sweden

M.A. Legal Philosophy, Zhongnan University of Economics and Law, China

Exchange graduate study, Criminal Justice, San Diego State University, U.S.

LL.B. Law, Zhongnan University of Economics and Law, China
2
Research focus
Main research areas:
 Environmental criminology
 Crime analysis and prevention
 Policing
Other research areas:
 Legal research in criminal justice
 Criminal justice education
 Death penalty
3
Projects

2013-2016 Co-PI, Ministry of Public Security of China. National Key Project Program (with
State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing,
Wuhan University, China). RMB 1,600,000 (about $ 261,200).

2013-2017 PI, Department of Education of Hubei Province, China. Chutian Scholars
Program. RMB 250,000 (about $ 41,000).

2013-2015 Co-PI, Ministry of Science and Technology of China. National Key Technology
Support Program 2012BAH35B03 (with State Key Laboratory of Information Engineering in
Surveying, Mapping and Remote Sensing, Wuhan University, China). RMB 10,000,000 (about $
1,632,250).

2013-2016 PI, Ministry of Education of China. Fundamental Research Funds for the Central
Universities 2722013JC030. RMB 30,000 (about $ 5,000).

2012-2015 PI, University Research Center Program. Ministry of Education of China.
Technology Improvement Project 31541210702, RMB 30,000 (about $ 5,000).

2010 Summer Research Fellowship, College of Criminal Justice, Sam Houston State University,
USA. $6,000.

2009 Summer Research Fellowship, College of Criminal Justice, Sam Houston State University,
USA. $6,000.
4
Research goal

Study crime and policing behavior at different space-time scales

Examine crime patterns and mechanisms

Integrate theory, methods, and application
5
Analytical methods
1. Temporal analysis of crime trend
2. Spatial analysis of crime and policing implementation
2.1 spatial analysis of crime pattern
2.2 hot spots policing research
3. Space-time analysis of crime and policing dynamics
3.1
3.2
3.3
3.4
3.5
repeat and near-repeat analysis
space-time interaction analysis
policing effect on near-repeat crime
cross-event space-time analysis
space-time chain analysis of crime
4. Regression analysis considering social and environmental factors
4.1 spatial panel regression
4.2 street network and Poisson regression
6
Study settings
Wuhan, China (2013 Census):
• 10.22 million people
• 8,494 km2 land area
• 1,203 persons/ km2
Houston, U.S. (2013 Census):
• 2.19 million people
• 1,553 km2 land area
• 1,431 persons/ km2
1. Temporal analysis of crime trend
Time-series analysis of burglary data in Wuhan, China (Exponential Smoothing)

Burglary
Number of events
Black: average
Red:2010
Green: 2011
Time (Day)
8
Software application
9
2. Spatial analysis of crime and policing implementation
10
2.1 Spatial analysis of crime pattern

Kernel density estimation of burglaries in Wuhan, 2013
11
2.2 Hot spots policing research

Texas Major City Research Initiative focuses on comparative data analysis and
evaluation of crime reduction interventions among “Big Six” police agencies in Texas.

Houston enhanced action patrol: Examining the effects of differential deployment lengths with a
switched replication design. Justice Quarterly, 2014.

The effects of gun possession arrests made by a proactive police patrol unit. Policing: An
International Journal of Police Strategies & Management, 2012.

Dallas’ disruption unit: Efficacy of hot spots deployment. Policing: An International Journal of
Police Strategies & Management, 2012.

Geographic information system effects on policing efficacy: A systematic review. International
Journal of Applied Geospatial Research, 2013.
12
2.2 Hot spots policing research
Houston Example: The study used a switched replication design
with repeated interventions in order to determine the dosage and duration necessary
to achieve reasonable crime reduction by rotating directed patrol units.
13
3. Space-time analysis of crime and policing dynamics
14
3.1 Repeat and near-repeat crime pattern
 The analysis examines patterns of gun assaults at the city-level as well as more localized levels in order
to understand the spatial distribution of near-repeats with in the city .
 Publication: Patterns of near-repeat gun assaults in Houston. Journal of Research in Crime and
Delinquency, 2012.
15
3.1 Repeat and near-repeat crime pattern
• Shenandoah, TX has been experiencing a fast growing economy and annexation
process. Auto burglary is a rising concern for small cities that rely highly on revenues
from malls and shopping. The research describes the temporal trends of auto burglary
offenses and compare hot spots of auto burglary offenses in different time periods.
 Publication: Space-time analysis of auto burglary patterns in a fast-growing small
city. International Journal of Applied Geospatial Research, 2012.
16
Software application
17
3.1 Repeat and near-repeat crime mechanisms

Publication: Space-time interaction of residential burglaries in Wuhan, China. Applied
Geography, 2014.
18
3.2 Space-time interaction analysis of crime
D(h, t ) 
K ST (h, t )  K S (h) KT (t )
K S (h) KT (t )
 Publication: Space-time interaction of residential burglaries in Wuhan, China. Applied
Geography, 2014.
19
3.3 Policing effect on near-repeat crime
 Publication: Proactive policing effects on repeat and near-repeat shootings in
Houston. Police Quarterly, 2011.
20
3.4 Cross-event space-time analysis



We developed AB Calculator Toolbox to study the cross-event space-time interaction, such as that
between burglary and auto-theft and that between crime and policing behaviors.
Publication: Deterrence or escalation: A micro-level analysis of firearm arrests’ effects on gun
violence in Houston, Texas. Paper under review.
Publication: Police response to repeat and near-repeat shootings in Houston, Texas. Paper under
review.
21
3.5 Space-time chain analysis of crime
(b)
(a)
3.0
15
amplifier
(in-degree < out-degree)
path
(in-degree = out-degree)
2.0
2.0
1.5
1.5
1.0
1.0
15
sink
(in-degree > 0, out-degree = 0)
bottle neck
(in-degree > out-degree)
5
15
10
isolate
(in-degree = out-degree = 0)
10
out degree
source
(in-degree = 0, out-degree > 0)
adjusted obs exp ratio
2.5
2.5
out degree
10
5
5
0
0.5
in degree
0
0
0.0
0
5
10
15
in degree
 Near repeat burglary chains: describing the physical and network properties of a
network of close burglary pairs. Dr. Michael Townsley, Griffith University
 Publication: Physical and network properties of a network of crime events:
Computation and implementation in policing practices. Paper under writing.
22
Software application
23
4. Regression analysis considering social and environmental factors
24
4.1 Spatial panel regression
 It integrates space and time in testing Social Disorganization Theory by examining relationship
between homicide rate and socioeconomic factors at community area level in Chicago,1960-1995.
 Publication: Analyzing the dynamics of homicide patterns in Chicago: ESDA and spatial panel
approaches. Applied Geography, 2011.
25
4.2 Street network and crime

It examines residential
burglaries at the street
segment level with
considerable variation
in street network
configuration and
socioeconomic factors.

Street permeability and
residential burglaries in
urban China. Applied
Geography, 2014.
26
Other research

Publication: A comparison of criminal justice education between U.S. and China:
An analysis of faculty’s characteristics. Theoretic Observation, 2014.

Publication: Visualizing research domains of criminology and criminal justice in
U.S. Higher Education Review, 2015.

Publication: The constitutionality of police strip searches in U.S. Journal of
Comparative Law, 2015.

Publication: A framework of quantitative research on judicial decisions: A study
of strip search lawsuits in U.S. (under review)

Publication: The policy and practice of death penalty without immediate execution
in U.S. and China. Jilin University Journal Social Sciences Edition, 2014.

Publication: The role of social media in the discussion of wrongful executions in
China. (under writing)

Publication: Lethal injection policy and practice in U.S. and China: A comparative
perspective. (under writing)

Publication: Gender and victimizations in public space: An integrative theoretical
perspective. Human Geography, 2015.
27
The next step

Predictive policing (e.g. PredPol software)

PredPol’s cloud-based software provides each law
enforcement agency with customized crime predictions for
the places and times that crimes are most likely to occur.
PredPol automatically 500*500 feet boxes for each shift of
each day.
28
Acknowledgement

Dr. William Wells, Sam Houston State University

Dr. Larry Hoover, Sam Houston State University

Dr. Michael Vaughn, , Sam Houston State University

Dr. Xinyan Zhu, Wuhan University, China

Dr. Xinyue Ye, Kent State University
29
Acknowledgement

Police Research Center, SHSU, TX

Law Enforcement Management Institute of Texas, TX

Crime Victims’ Institute of Texas, TX

Institute for Legal Studies in Criminal Justice, SHSU, TX

Houston Police Department, TX

Dallas Police Department, TX

Shenandoah Police Department, TX

Wuhan City Police Department, China

State Key Lab of Information Engineering in Surveying, Mapping and Remote
Sensing, Wuhan University, China

Wuhan Prosecutor’s Office, China

Terrorism Research Center, ZUEL & Xinjiang Police College, China
30
Thanks!
31
Download