Crime and spatial structure of cities.

Saraswati Jigajinni
Abstract:

Today cities are one of the most important and integral part of our lives. This is where
we spend our lives. Even if we are in the house inside a room we still are connected to
the city activities. Sense of Feeling Safe and unsafe in the cities matter. The components
that makes one feel safe may differ from one person to another and also depends on one
place to other. It could be density of space, population density, Area, landscaping, traffic,
Crowd around, Lights, Security guard, or could be any of the element of the city. One may
feel threatened in a crowd and someone else could feel safe in the same crowd.
Accourding to New criminology study criminals tend to commit crimes in areas they are
familiar with. In this research paper we will look at the criminal activity structures that is
represented by spatial network of sites.

Keywords:, crime mapping, crime and city, social disorganization, crimes spatial
structure. Crime and network of sites

Introduction

Understanding crimes and the organisational process that runs in criminals mind. And the
spot where the crime could happen and the regions of the city that can lead to the better
planning of crime is important to identify the network of the criminal activities. For
instance, Information of crime is always attached by location of the crime, metadata which
indicates where exactly the criminal activity has occurred. This spatial data play crucial
role in understanding criminal behaviour in olden days. And many such areas are refereed
as hostspots and many theories have proposed to understand and explain the crime areas.
The study helps us to Recognizing the activity role in a area. For instance, Many of
community aspects and neighbourhood activities leads to inability. It is difficult to control
the behaviour of the people in a community or neighbourhood and that can lead to crime.
Also it is also evidently studied that crimes can happen more with people who stay in
similar geographical area or who are familiar with each other. Hotspot and histogram work
similarly. In sub-regions Histogram allows one to depict crime frequency which is similar
measure.

In order to capture this structure, we propose to analyse crime activity using a network
framework. The rationale here is that criminal activities rely heavily condifferent types of
networks, such as a social network within gangs and the word-of-mouth flow of
information regarding illegal market. The presence of these relations suggests that network
science can be a powerful tool for analysing criminal activity. In this paper, we define a
crime network in which crimes are connected if they occur nearby. We used data of crime
incidents from US police department records to generate complex networks of criminal
activity in Los Angeles, CA, and Miami, FL and we found borders between communities
in these structures using an approach adapted from Thiemann et al. [16]. Finally, in order
to show that these networks capture real-world phenomena, we compared these borders to
similar boundaries in demographic data. We found statistically significant variations that
 Author’s name / Insert your paper’s title here

indicates the structure of the crime networks reflects real-world phenomena that cannot be
accounted for by demographic differences.

Types of crimes in koramangala are Rape, Dowry, Domestic violence, Illegal drug trade,
Arms trafficking, Petty crime, Confidence tricks

Commercial zone : The crimes like Theft, Larceny, Robbery, are very commonly
happening in this area.There are lots of shops,pubs and resturants in this area and hence
the moment of people visibly seen. Also one of the reason is the people visting restaurants
and pubs leave the places late and some times they are alone, or two. This makes the thief
easy to commit crimes.

Low Income settlement and Residential Areas : The security and violence ,Abusing is a
type of crime that is seen in this area. people here are not educated, There day today needs
force them to commit crime. And for some abusing women, kids or eldely people in the
family is like a routine.

Most of the crimes in Koramangala Happens due to Population density and degree of
urbanization.

Variations in composition of the population, particularly youth concentration.

Stability of the population with respect to residents’ mobility, commuting patterns, and
transient factors.

Modes of transportation and highway system.

Economic conditions, including median income, poverty level, and job availability.

Cultural factors and educational, recreational, and religious characteristics.

2
Family conditions with respect to divorce and family cohesiveness.

Climate.

Effective strength of law enforcement agencies.

Administrative and investigative emphases of law enforcement.

Policies of other components of the criminal justice system (i.e., prosecutorial, judicial,
correctional, and probational).

Citizens’ attitudes toward crime.

Crime reporting practices of the citizenry.

criminalactivity typically occurs in the sphere of familiarity of the criminal. Despite

thissphere of familiarity being peculiar to the individual, areas of high traffic, suchas
downtown areas, lie within the sphere of familiarity of many individuals. Thisaspect is
also related to the fact that criminals typically commit crimes withina short distance from
their home.

Metropolitan areas are typically organized by regions of different land usethrough the
natural development of urban centers and unnaturally through zon-ing laws. These
different land uses include: residential use, commercial use, andindustrial use. The
presence of types of crimes differs between these land uses neighborhoods with residential
housing and no commercial businesses are per-ceived as safe and non-residential land uses
are correlated with an increase incriminal activity . Non-residential areas is typically
found to have higher traf-fic in comparison to residential areas, consequently they witness
to more crime Also, the places within these areas, such as shopping centers or public
parks, coincide with an increase in foreign or non-residential presence. The presence of
such strangers negatively impacts a neighborhood’s social structure.

Building Networks and the Borders

The domestic Crime rate is also high in the area where the low income people are more
and below are marked the low income middle income and high income areas of
koramangala.
 Author’s name / Insert your paper’s title here

Street networks influence human mobility and thus the potential offender smobility. In
fact, the convergence of a potential offender with a criminal oppor-tunity is much more
likely to occur and be exploited on a street that is relatively accessible and frequently
traveled . Street networks are not only correlated with an increase in crime incidence but
additionally have a relationship with the typical journey-to-crime length of an offender .
Roadways and public transportation link together different areas of a criminal’s sphere of
familiarityand facilitate travel outside of a criminal’s immediate neigh borhood. The type
of crime can affect the journey-to-crime length. For example, violent crime tripsare shorter
in length than property crime trips .Both of aforementioned theories serve to explain
crimes in terms of the spatial context of the neigh borhood. By the social disorganization
theory, a break down in the social structure of a neigh borhood causes the elevated crimes
rates. On the other hand, the routine activity theory implies that the higher traffic
associated with non-residential land use is the cause of such crime rates. The choice of the
location information that each vertex represents also defines the resolution of the network. he
comparison between different sets of borders for the same region is made with the absolute cross
correlation of a network representation of the borders. A set of borders can be embedded in a
network by representing the physical regions as vertices, and the borders between them as edges
connecting adjacent regions. These edges have weights representing the strength of the borders be-
tween the regions. The absolute cross correlation is the normalized scalar product of the weights of
edges in two networks.

A lack of correspondence between the socio-demographic characteristics and the borders of crimes
networks suggests that such structures are not just aproxy for the characteristics of the an area. But
before we assume that the borders structures are representing some underlying network phenomena,
weneed to validate it against a null model. In order to carry this analysis out, wechose to generate
random borders based on the demographic borders and reuse these for the comparison to each set of
crime borders.

4
 In order to test whether the borders of crimes were capturing some underlying network structure we
compared them against borders generated by demographic data. To ensure the comparison reflects
the location, not the number of borders ,we needed the demographic data and the crime data to have
a similar number of communities. To get borders in the demographic data of roughly the same
resolution as the borders in the crime networks we cut the dendrogram of demo-graphic clusters at
four heights resulting in a similar number of communities in both data sets. The three methods of
hierarchical clustering and the four levels we cut each dendrogram of demographic borders to com-
pare each network to. The clusters for single linkage tended to be much more sparse than the other
two methods of clustering so we dropped all but the most fine resolution of borders for single
linkage. This left us with nine sets of borders for each metropolitan area we analyzed.

Conclusions

The Spatial Structure of Crime in Urban Environments crime networks could be used to find
interesting patterns in the structure of criminal activity in a metropolitan area. Conclusions and
Future Works The reliance of criminal activities on different types of networks makes network
science an obvious choice for analyzing criminal activity. Using assumption sabout the spatial
distribution of crimes, we proposed a new model in which networks are built with links between
crimes in close physical proximity. Weshowed that there exists an underlying structure of criminal
activity that can be represented by the spatial distribution of the crimes. The model we propose
allows the construction of criminal networks without the need for gathering ex-tra data about
individual criminals or patterns of crimes. No additional work is required to gain insights available
from this new model because the data we used to construct the networks is already a part of law
enforcement bookkeeping. The growth of computational resources available to law enforcement
agencies allowsfor more complex analysis than the heat maps that have been useful in the past. we
generated networks for multiple spatial distances, ranging from 0.1 miles to 3.2 miles, and various
crime types. We showed that these networks capture real-world phenomena, by comparingborders
between communities of crimes. We compared these borders to similarboundaries in demographic
data and found statistically significant variations, which indicates the structure of the crime networks
is reflecting real-world phe –nomena

Understanding crimes and the organisational process that runs in criminals mind. And the spot where
the crime could happen and the regions of the city that can lead to the better planning of crime is
important to identify the network of the criminal activities. Crimes can be prevented and some of the
simple steps that are mentioned can prevent crime and keep us safe.

When you are looking of crime prevention to improve security around, the best way to approach it is
to look at premises as if you were the offender. Identify the weak spots, vulnerable areas and
concealment points and prioritise the areas for improvement. Contact our Crime Prevention Officers
before you undertake any improvements and they will work with you to ensure that you are taking
the best approach possible for your respective circumstances.