Вы находитесь на странице: 1из 16

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/309125859

Crime Mapping and Analysis using GIS

Technical Report · July 2016


DOI: 10.13140/RG.2.2.11064.14081

CITATIONS READS

0 11,638

1 author:

Pranav Kedia
International Institute of Information Technology Bangalore
1 PUBLICATION   0 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Emergency Vigilance System View project

All content following this page was uploaded by Pranav Kedia on 13 October 2016.

The user has requested enhancement of the downloaded file.


Crime Mapping and Analysis using GIS
Pranav Kedia
Student, International Institute of Information Technology, Bangalore

Abstract:
This paper looks at the possibility of utilizing GIS for crime mapping and its analysis for
effective law enforcement and crime management. This option was explored by showing the
procedural method of creating a) A hierarchal system of location data collection which
previously didn’t exist b) Creating database with all spatial and non-spatial data c) Creating
heat map based on temporal and spatial analysis such as query and KDE using QGIS
software and GPS. The analysis shows area of crime hotspots, areas deficient in security
outfit, areas requiring constant police patrol, time and days when citizens face the maximum
crime and of what type. The study proves that GIS gives us a better synoptic perspective to
crime mapping, study, analysis, decision making, better presentation to a layman and thus
prevents crime. It however suggests that migrating from traditional method of crime
management to GIS demands building the required infrastructure and recruiting trained
personnel and facilities.

Keywords: GIS, Hotspot, Geospatial Analysis, KDE (Kernel Density Estimation), Temporal
Analysis, Aoristic Analysis.

1. Introduction:
Crime is not spread evenly across maps. It clumps in some areas and is absent in others.
People use this knowledge in their daily activities. They avoid some places and seek out
others. Their choices of neighbourhoods, schools, stores, streets, and recreation are
governed partially by the understanding that their chances of being a victim are greater in
some of these places than in others. In some places people lock their cars and secure
belongings. In other places they do not. Along some streets people walk swiftly and view
approaching strangers with suspicion. Along other streets they casually stroll and
welcome the next interesting person they might meet, and notice others making the same
choices in the same areas.
Some might argue that this behaviour merely shows that people are
unreasonably fearful of some areas but not of others. This may often be true, but the fact
that people are not equally fearful of all places suggests that they understand that crime is
not evenly distributed. People might be mistaken about the risks of some places, but they
are not mistaken that their risk of being a victim of crime is not geographically constant.
In India, Police organisation plays a critical role. With the increase in crime in recent
past, police should use current technologies to give themselves the much needed edge.
But still worse is the fact that Law enforcement agencies in most developing
nations are yet to be computerised for effective record keeping, analysis of cases, easy
reference and retrieval, storage of information which in future would provide reliable and
comprehensive data round the clock, help in trend prediction and give decision support to
Police agencies. As an entity crime has spatial attributes like time, location and process
(Under section). Thus, if this spatial information about crime prone areas is available
timely available and with quick ease then it will help in effective policing of the state.
But with the old systems in use the cops seems to be lagging behind the sophistically
growing criminals in the eternal race of law breakers and law enforcers.
With police departments of most developing nations using analog and outdated filing
systems in most cases, the police operations are carried out by intuition, tip-off
information or simple ‘trial-error’ method. But as command staff need to make informed
decisions, GIS gives you the ability to improve planning, reduce costs, and increase
transparency. With GIS, your department can apply the power of geography to organize,
visualize, and disseminate mission-critical data. From the command center to officers in
the field, GIS gives your entire organization the ability to improve planning, operational
awareness, and field mobility. You can enhance response times, coordinate enforcement
activities, and better understand your jurisdiction. With the GIS platform, your
department can derive meaning from large volumes of data so that crime patterns—and
the appropriate policing strategies—become evident. This study therefore seeks to
explore the capability of GIS in crime mapping using Faridabad District of Haryana as
case study.

2. Study Area:
Faridabad is the largest city in the north Indian state of Haryana in Faridabad district. It
is a leading industrial centre and situated in the National Capital Region bordering the
Indian Capital New Delhi. It is surrounded by Delhi to its North, Noida to its North
East, Greater Noida to its East and Gurgaon to its West. The river Yamuna forms the
eastern district boundary with Uttar Pradesh. Government of India included it the second
list of Smart Cities Mission on 24 May 2016.
Based on historical, political, economic, educational, administrative and nodal attributes,
Faridabad has witnessed tremendous growth in size, population and Per Capita Income
which in turn has simultaneously led to the growth in crime. The crime here exists in
many forms major amongst them being Crime against women, murders, theft, illegal
drug and liquor trade and its consumption in public. Also, it is no longer a hidden fact
that cities provide for a conducive setting and breeding ground for organised and
specialised crime. Various social factors such as increasing unemployment, increasing
inequality based on wealth have brought alienation between the have and the have-nots,
thus accelerating urban crimes in recent times. Despite their best effort the law
enforcement agencies are lagging behind the criminals who are becoming sophisticated
on daily basis. As a result, citizens tussle everyday with fear of unknown and insecurity
of their life and property. Unfortunately, even today there is a little or no application of
the primitive pin-on-map system in some stations of the city, let alone the use of GIS. In
the age of Information Communication Technology, the non-application of GIS and geo-
database is costly and counterproductive in the fight against crime.

3. Conceptual Framework:
Crime theories are critical for useful crime mapping because they aid in the interpretation
of data and provide guidance as to what actions are most appropriate. Therefore,
understanding how crime theories account for hot spots is critical. Several theories of
crime and disorder concentration (hot spots) exist. Some theories disagree, but often the
theories do not contradict each other. Rather, they explain different types of crime
phenomena that occur at different geographic levels.
Each level has basic units of analysis—the things being examined. One can think of units
as corresponding to the geographic areas being depicted on maps: points, lines, or
polygons. Some theories help explain point concentrations of crime. Other theories help
explain linear concentrations of crime or hot spot crime polygons. However, theories of
crime are useful for helping to guide crime and disorder mapping only if one selects a
theory appropriate for the level of analysis and action.
These theories include:

Place theories which tend to explain why crime events occur at the lowest level of
analysis-specific places, usually these places are points may be a street corner, or maybe
other very small places.
Street theories which tend to deal with crime that occur at a slightly higher level than a
specific place, maybe small, stretched areas like blocks or streets.
Neighbourhood theories which tend to explain neighbourhood differences i.e. why not
all neighbourhoods have same level of crime, what areas are claimed by gangs and mobs.
Repeat Victimisation theories which deal with why same victims are targeted repeatedly.

Then we come to the different types hotspots and crime theories associated with them

Repeat places hotspot is the most basic form of hotspot is a place that has many crimes.
The theory associated with is routine activity theory helps to explain why crime often is
concentrated at some places. It particularly points how behaviour is regulated by the
managerial staff. For example, the difference between a bar with no incidents and a bar
with many incidents or assaults is most likely to be the regulation of first bar’s
employees over the drinking limit of their patrons thus minimizing the chances of any
incident or assault, attracting who like a regulated place and repelling people who want
an unregulated place.
Repeat Victimisation Hotspots refers to multiple attack on same individual, regardless of
location
The theory associated with has got much less role of managerial staff. In those cases, one
should look for occupation, lifestyle and commuting patterns of the potential victim
(Farrell and Pease, 1993; Spelman, 1995b; Stedman and Weisel, 1999). As the most
likely these people are most likely indulged in deviant or criminal activity (e.g., drug
dealing, drug use, heavy alcohol consumption, prostitution) (Menard, 2000). Even some
occupations increase the chances of repeat victimisation like police officers have a higher
rate of victimisation than many other occupations.
Repeat streets hotspots deals with thoroughfares and streets with a high degree of
victimisation.
The offender finds targets while going from their legitimate business – going to and from
work, recreation, shopping, school or any daily routine activity. Potential targets that are
not along or close to offenders’ routes will unlikely be victimised but those along or
close to offenders’ routes have an elevated risk of victimisation. Also Thoroughfares
concentrate people, so these targets that are situated along thoroughfares face higher
crime risks than targets away on side streets far from thoroughfares (NIJ 2005).
Neighbourhood and other hotspots
Most is written in the subject of neighbourhood crime theories in Social Factors in
juvenile delinquency (1931) Shaw and McKay wrote about the unchanged state of some
neighbourhoods regardless of time and who lived there. Some talk about how local
residents have the ability to control this deviancy (Bursik and Grasmick ,1933).
Social disorganisation Theory talks about influence of social networks in
preventing deviancy and how in certain neighbourhoods the constant turnover and
outmigration, disrupts social networks or prevents them from forming.
Social efficacy which is “the willingness of locals to intervene for common
good” also influences deviancy. The neighbourhoods with a great deal of social efficacy
have less crime than the neighbourhoods with low levels.
Broken windows theory (Wilson and Kelling 1982) tells us how small
transgressions of social norms result in social pressures to bring the offending party into
compliance. Once a place becomes untended, it undermines the ability and willing ness
of residents to enforce social order. Thus, they withdraw from these norms and let further
deviancy to occur.

4. Methodology:
Data Collection:
The biggest problem that we faced was that the study area had no provision for collecting
location (coordinates) of the place where crime was committed. So, as far as our initial
study was concerned the data we collected was through using location sharing feature of
WhatsApp, which was sent by the concerned Investigation Officers to Our operators
which would file the location into different spreadsheets based on Police Station Area
along with all other non-spatial attributes like FIR (First Information Report) No., Time
of crime, Section of IPC (Indian Penal Code) under which it was registered, Physical
address. Hence, through this hierarchal system the interim problem of data of crime was
solved.

Maps:
For our analysis we used maps form Open street maps provided as plugin to our GIS
Software QGIS.

Ground Investigation:
For the purpose we used a handheld high precision GPS (GARMIN GPSMAP 64s
handheld GPS) and a Digital Camera for pictorial coverage of hotspot (where possible).

Data Collection

Plotting on Map

Analysis
5. Process of Hotspot Identification:
5.1 Spatial hotspots:
With the advent if IT (Information Technology) age, we have been able to apply some
spatial techniques which include the use of location quotients (Brantingham &
Brantingham, 1995b), the development of kernel surface estimation algorithms
(McLafferty, Williamson, & McGuire, 2000), and Local Indicators of Spatial
Association (LISA) such as Getis and Ord Gi and Moran Local I (Mencken & Barnett,
1999; Messner et al., 1999). Some spatial Analyses uses an underlying social
phenomenon to find the rate of crime which necessitates the use of boundary units, such
as census tracts and police beats. But some researchers have also recognised that a
simple spatial concentration of crime can be valuable.
A no. of other programs is available which will not give a specific shape but would
produce a hotspot surface map called heatmap rather similar to the temperature one
might sight on the weather report (McLafferty et al., 2000). These map actually show
kernel density surface map of the crime intensity (Rosenblatt, 1956 and Parzen, 1962),
this process involves estimating the density of crime across an entire two dimensional
study area, based on the known locations of discrete events. KDE begins with overlaying
grid on top of study area and calculating a density estimate based on the center points of
each grid cell. Each distance between an incident and the center of a grid cell is then
weighted based on a specific method of interpolation (e.g. the kernel function) and
bandwidth (Hart and Zandbergen, 2013). Figure 1 illustrates the KDE process and shows
a number of parameters that must be considered before a density estimate can be
produced. These parameters include the grid cell size, the method of interpolation, and
the bandwidth. Here, Figure 2 illustrates an actual KDE heat map of Area under Central
police station in the study area and it includes Crime data of all crimes (Robbery, Crime
against women, Corruption, Selling Illegal drugs and liquor etc.) for a period of Two
months i.e. june’16 and july’16.

Figure 1: An illustration of kernel density estimation


Courtesy: Hart and Zandbergen, 2013
Figure 2: A KDE heatmap of area under central police station of the study area. Here deep red is colour for highest
intensity, then light red, followed by yellow and least intensity is for blue colour

5.2 Temporal Hotspots:


In comparison to spatial analysis much less effort has been given to temporal dynamics
of local crime patterns (Ratcliffe, 2004). Some works have looked for patterns over
lengthy period of time, over have studied seasonality of crimes and some have viewed
the risk of repeat victimisation throughout the day. But what appears to have left
everybody stuck is that the source of this researches i.e. police crime databases. Given
that there is latency involved in the police reaching a crime scene, also when a property
theft occurs the police who are rarely at the crime scene so they are left to record a start
time (time it was last seen) and end time (time it was found missing). Ratcliffe and
McCullagh (1998a) initially proposed the development of the temporal weighting to
include the spatial component, an idea further refined for analytical and mapping purpose
and referred to as ‘Aoristic analysis’. This process is explained in figure 1 with an
example.
Courtesy: Ratcliffe, 2004, The hotspot matrix: A framework for the spatio-temporal Targeting of crime reduction

6. Results and Discussion:


Now that we have laid some conceptual foundation for our spatio-temporal mapping,
we would like to see what were the results?
6.1 Application advantages:
Using GIS, we were able to query data over all the spatial and non-spatial attributes.
Thus, with the help of our data we were able to lay a simple SQL query to data sheet
and obtain the resulting points in a different layer or just colour separated. This gave
us power to view data for temporal data (Time of the day and day of the week). The
structures thus obtained were helpful in the way that we realized that over a hotspot in
the study area most robberies were reported on Tuesday in the period of 2 months,
But the part that was missing was the why, so we went to the hotspot on a Tuesday for
our ground analysis, and we found that the weekly market at that place attracted a lot
of people and thus, provide for a conducive environment for the robbery especially
vehicle theft to take place. Figure 6 illustrates the map of the place under Police
Station Bhupani of the study area. Figure 7 shows the actual picture of scene.

Figure 6: Theft Under 379 IPC in Bhupani Figure 7: Ground Research Photo of the weekly Bazaar
Police station
Based on the temporal attributes of the time of the day it was observed that crime
especially associated with vehicle theft in most cases have reported time in ranges
from (time of last seeing) and to (time when found missing), same is the case for
kidnapping cases and burglary. Thus, we tend to us the hotspot matrix classification
given by Ratcliffe (2004) on hotspot matrix framework for spatio-temporal targeting
of crime reduction. This helps us to use maps and aoristic analysis. Now we will see
the types of hotspot based on the above classification.

Figure 7: Three Types of Spatial Hotspot: Dispersed (A), Clustered (B), and Hotpoint
(C). Each image shows 12 crime points located within a hotspot. In (A) the points are
dispersed throughout the hotspot. In (B) there are points throughout the hotspot but
there is clear evidence of clustering at one point in the top left of the hotspot. In (C),
all 12 points are co-located at the same point in the centre of a circular hotspot.
Courtesy: Radcliffe (2004), The hotspot matrix: A framework for the spatio-
temporal Targeting of crime reduction

Figure 8: Three Temporal Hotspot Categories. In each figure, the probabilistically


weighted offence times are shown as vertical bars. Each bar represents one hour,
running left to right from 00:00–00:59 to 23:00–23:59 hours, or more simply, hourly
from midnight to midnight. The three temporal hotspot types are diffused (A), focused
(B), and acute (C). In (A) it can be seen that while the risk of crime rises and falls
there is a general risk throughout the hours of the day and night. In (B), although there
is a general level of risk, there is also a period in the evening of significantly higher
crime risk. In (C) it can be seen that the risk of crime is most acute in the early
evening only.
Courtesy: Radcliffe (2004), The hotspot matrix: A framework for the spatio-
temporal Targeting of crime reduction

Let’s see these patterns in the real life scenario from the data we collected during our
study. These maps show the different hotspot for central police station area for theft
registered under section 379 IPC with graphs for aoristic value.
Figure 9,10: The above is point displacement map which show all overlapping points
on map with red dot as the specific coordinate dot and blue to suggest the no. of dots.
These two figures suggest that the hotspot is an acute, Hotpoint at a shopping centre
of sec-14, Faridabad, Haryana, India. Drawn from data of two months (May’16 and
June’16) it shows a hotspot dominated by the one at the shopping center which
suggests it is a Hotpoint. The histogram suggests that there is significant block from
12:00 to 18:00 hours where majority of crime is incident suggesting its acute nature.

6.2 Policing solution for hotspots:


There is also an understanding that certain policing tactics can influence the short-
term patterns of crime. Indeed, the use of local community partnerships can take some
time to coordinate and organize, and some law enforcement tactics can be
implemented at short notice to fill the gap. These include crackdowns (Sherman,
1990), intelligence-led operations (Ratcliffe, 2001), and problem-oriented policing
(Goldstein, 1990; Leigh, Read, & Tilley, 1996). These strategies are part of a whole
toolbox of possible tactics that an operational commander can employ, including
flooding an area with uniform patrols, deploying surveillance units, using car patrols,
and in some countries deterring offenders by setting up random breath tests in high
crime areas. Some of these tactics are geared more towards crime prevention, while
others (such as the use of surveillance units) have a crime detection focus. The table
gives all possible solutions based on the type of hotspot (temporally and spatially).

Policing Hotspot Matrix Spatial

Dispersed Clustered Hotpoint

Diffused Uniform vehicle Random breath Roadblocks,


Temporal patrols, tests, foot plain clothes
architectural patrols, patrols, random
changes, public architectural breath tests.
education changes, Private security,
campaign publicity CCTV
campaign
Focused Uniform vehicle Vehicle and Surveillance
and foot patrols, foot patrols, units, plain
improved random breath clothes, CCTV,
Lighting, public tests, private surveillance of
education security, entry/exit points
campaign improved
lighting
Acute Unmarked Surveillance Surveillance,
vehicle patrols, and plain arrest squads,
private security, clothes patrols, CCTV,
improved CCTV unmarked
lighting police units

6.3 Crime pattern shifts and reduction:


Our two months of data on crime in the study area gave us glimpse of how the crime
being a human phenomenon is not at all random and dispersed evenly over the whole
study area but it is concentrated for few spots (Hotspots) and that police measure and
patrol beats designed with the help these concentration regions make us shift the
Crime hotspot and reduce the intensity of crime in these regions. Figure 4,5 illustrates
how the crime shift appears on the map, How the intensity of crime over certain areas
can be reduced. Figure 4 is a heat map created with the crime data of one month i.e.
16th May’16 to 15th June’16. Figure 5 illustrates how Making changes based on the
hotspot and by doing a hotspot oriented policing we can reduce crime in the areas to a
measurable extent.
Figure 4: heatmap from crime data of central police station of study area

Figure 5: Heatmap from crime data of same police station depicting the shift and
reduction in crime.

The above comparison yields us with important information that GIS based heatmap
helps the commanding officers with support in decision making and help in effective
use of the available force and scarce resources.
References:
Brantingham, P. L. and Brantingham, P. J. (1995a) Criminality of place: Crime generators
and crime attractors, European Journal of Criminal Policy and Research, 3(3), 5–26.
Bursik, R.J., and H.G. Grasmick. 1993. Neighbourhoods and Crime: The Dimensions of
Effective Community Control. New York, NY: Lexington Books.
Corcoran J. J., Wilson I. D. & Ware J. A., 2003. Predicting the geo-temporal variations of
crime and disorder. International Journal of Forecasting 19, 623–634.
Conference on Geomatics in Electronic Governance, January 2000, Pune.

Eck John E., Chainey Spencer, Cameron James G., Leitner Michael, and Wilson Ronald E.
(2005), Mapping Crime: Understanding HotSpots, NIJ.

Farrell, G., and K. Pease. 1993. Once Bitten, Twice Bitten: Repeat Victimization and Its
Implications for Crime Prevention (vol. 46). London: Her Majesty’s Stationery Office.
Goldstein, H. (1990) Problem-oriented policing. New York: McGraw-Hill.
Gupta R., Rajitha K., Basu S., Mittal S.K., 2012. Application of GIS in Crime Analysis:
Gateway to Safe City.
Johnson, C.P., 2000. Crime Mapping and Analysis Using GIS. Geomatics 2000:
Leigh, A., Read, T. and Tilley, N. (1996) Problem-oriented policing., Police Research
Group: Crime Detection and Prevention Series, 75.
Menard, S. 2000. “The ‘Normality’ of Repeat Victimization from Adolescence Through
Early Adulthood.” Justice Quarterly, 17(3), 543–574
McLafferty, S., Williamson, D. and McGuire, P. G. (2000) Identifying crime hot spots using
kernel smoothing. In V. Goldsmith; P. G. McGuire; J. H. Mollenkopf, & T. A. Ross (Eds.),
Analyzing crime patterns: Frontiers of practice (pp. 77–85). Thousand Oaks, CA: Sage.
Mencken, F. C. and Barnett, C. (1999) Murder, non-negligent manslaughter and spatial
autocorrelation in mid-South counties, Journal of Quantitative Criminology, 15(4), 407–422.
Messner, S. F., Anselin, L., Baller, R. D., Hawkins, D. F., Deane, G. and Tolnay, S. E. (1999)
The spatial patterning of county homicide rates: An application of Exploratory Spatial Data
Analysis, Journal of Quantitative Criminology, 15(4), 423–450.
Ozkan, K., 2004. Managing data mining at digital crime investigation. Forensic Science
International 146, S37-S38.
Patricio Tudela., 2004. Trends in Crime Mapping: The Challenges and Perspective of
Integrating Crime Mapping in Local Safety and Governance Policies. CPRD.
Ratcliffe, J. H. and McCullagh, M. J. (1998a) Aoristic crime analysis, International Journal
of Geographical Information Science, 12(7), 751–764.
Ratcliffe Jerry H.,2004. The Hotspot Matrix: A Framework for the Spatio-Temporal
Targeting of Crime Reduction.

Spelman, W. 1995b. “Once Bitten, Then What? Cross-Sectional and Time-Course


Explanations of Repeat Victimization.” British Journal of Criminology, 35, 366–383.

Stedman, J., and D.L. Weisel. 1999. “Finding and Addressing Repeat Burglaries.” In C.S.
Brito and T. Allen (eds.), Problem-Oriented Policing: Crime- Specific Problems, Critical
Issues and Making POP Work (vol. 2, pp. 3–28). Washington, DC: Police Executive
Research Forum.

Sherman, L. W. (1990) Police crackdowns: Initial and residual deterrence. In M. Tonry & N.
Morris (Eds.), Crime and justice: An annual review of research (pp. 1–48). Chicago:
University of Chicago Press. Ratcliffe, J. H. (2001) Policing urban burglary., Trends and
Issues in Crime and Criminal Justice,213.

Shaw, C.R., and H.D. McKay. 1931. Social Factors in Juvenile Delinquency (vol. 2, no. 13).
Washington, DC: U.S. Government Printing Office.

Timothy Hart, Paul Zandbergen, (2013) Kernel density estimation and hotspot mapping.

Toju Francis Balogun, Henry Okeke, Christain Ifanyin Chukwukere (2014), Crime Mapping
in Nigeria Using GIS, Journal of Geographic Information System, Scientific Research
Publishing Inc.

Wilson, J.Q., and G.L. Kelling. 1982. “Broken Windows: The Police and Neighbourhood
Safety.” Atlantic Monthly, March, 29–38.
View publication stats

Вам также может понравиться