Crime Analysis using K-Means Clustering

Abstract

today's world security is an aspect which is given higher priority by all political and government worldwide and aiming to reduce crime incidence. As data mining is the appropriate field to apply on high volume crime dataset and knowledge gained from data mining approaches will be useful and support police force. So In this paper crime analysis is done by performing k-means clustering on crime dataset using rapid miner tool.

Full text

International Journal of Computer Applications (0975 – 8887)
Volume 83 – No4, December 2013
1
Crime Analysis using K-Means Clustering
Jyoti Agarwal
Mtech CSE
Amity University,Noida
Renuka Nagpal
Assistant Professor
Amity University ,Noida
Rajni Sehgal
Assistant Professor
Amity University,Noida
ABSTRACT
In today’s world security is an aspect which is given higher
priority by all political and government worldwide and aiming
to reduce crime incidence. As data mining is the appropriate
field to apply on high volume crime dataset and knowledge
gained from data mining approaches will be useful and
support police force. So In this paper crime analysis is done
by performing k-means clustering on crime dataset using
rapid miner tool.
Keywords
Cluster, Crime Analysis and Rapid miner
1. INTRODUCTION
In present scenario criminals are becoming technologically
sophisticated in committing crime and one challenge faced by
intelligence and law enforcement agencies is difficulty in
analyzing large volume of data involved in crime and terrorist
activities therefore agencies need to know technique to catch
criminal and remain ahead in the eternal race between the
criminals and the law enforcement. So appropriate field need
to chosen to perform crime analysis and as data mining refers
to extracting or mining knowledge from large amounts of
data, data mining is used here on high volume crime dataset
and knowledge gained from data mining approaches is useful
and support police forces. To perform crime analysis
appropriate data mining approach need to be chosen and as
clustering is an approach of data mining which groups a set of
objects in such a way that object in the same group are more
similar than those in other groups and involved various
algorithms that differ significantly in their notion of what
constitutes a cluster and how to efficiently find them. In this
paper k means clustering technique of data mining used to
extract useful information from the high volume crime dataset
and to interpret the data which assist police in identify and
analyze crime patterns to reduce further occurrences of similar
incidence and provide information to reduce the crime. In this
paper k mean clustering is implemented using open source
data mining tool which are analytical tools used for analyzing
data .Among the available open source data mining suite such
as R, Tanagra ,WEKA ,KNIME ,ORANGE ,Rapid miner.k
means clustering is done with the help of rapid miner tool
which is an open source statistical and data mining package
written in Java with flexible data mining support options. Also
for crime analysis dataset used is Crime dataset an offences
recorded by the police in England and Wales by offence and
police force area from 1990 to 2011-12 .In this paper
homicide which is crime committed by human by killing
another human is being analyzed .
This paper is divided into 7 sections: Related work, Proposed
System Architecture, Experimental set up & Results,
Conclusion, Future scope, References
1.1 Crime analysis
Crime analysis is defined as analytical processes which
provides relevant information relative to crime patterns and
trend correlations to assist personnel in planning the
deployment of resources for the prevention and suppression of
criminal activities
It is important to analyze crime due to following reasons :
1. Analyze crime to inform law enforcers about general and
specific crime trends in timely manner
2. Analyze crime to take advantage of the plenty of
information existing in justice system and public domain.
Crime rates are rapidly changing and improved analysis finds
hidden patterns of crime, if any, without any explicit prior
knowledge of these patterns.
The main objectives of crime analysis include:
1. Extraction of crime patterns by analysis of available
crime and criminal data
2. Prediction of crime based on spatial distribution of
existing data and anticipation of crime rate using
different data mining techniques
3. Detection of crime
2. RELATED WORK
Data mining in the study and analysis of criminology can be
categorized into main areas, crime control and crime
suppression. De Bruin et. al. [1] introduced a framework for
crime trends using a new distance measure for comparing all
individuals based on their profiles and then clustering them
accordingly. Manish Gupta et. al. [2]. highlights the existing
systems used by Indian police as e-governance initiatives and
also proposes an interactive query based interface as crime
analysis tool to assist police in their activities. He proposed
interface which is used to extract useful information from the
vast crime database maintained by National Crime Record
Bureau (NCRB) and find crime hot spots using crime data
mining techniques such as clustering etc. The effectiveness of
the proposed interface has been illustrated on Indian crime
records. Nazlena Mohamad Ali et al.[3] discuss on a
development of Visual Interactive Malaysia Crime News
Retrieval System (i-JEN) and describe the approach, user
studies and planned, the system architecture and future plan.
Their main objectives were to construct crime-based event;
investigate the use of crime based event in improving the
classification and clustering; develop an interactive crime
news retrieval system; visualize crime news in an effective
and interactive way; integrate them into a usable and robust
system and evaluate the usability and system performance and
the study will contribute to the better understanding of the
crime data consumption in the Malaysian context as well as
the developed system with the visualization features to
address crime data and the eventual goal of combating the
crimes .Sutapat Thiprungsri [4] examines the application of
cluster analysis in the accounting domain, particularly
discrepancy detection in audit. The purpose of his study is to
examine the use of clustering technology to automate fraud

International Journal of Computer Applications (0975 – 8887)
Volume 83 – No4, December 2013
2
filtering during an audit. He used cluster analysis to help
auditors focus their efforts when evaluating group life
insurance claims. A. Malathi et al.[5] look at the use of
missing value and clustering algorithm for a data mining
approach to help predict the crimes patterns and fast up the
process of solving crime. Malathi. A et. al.[6] used a
clustering/classify based model to anticipate crime trends. The
data mining techniques are used to analyze the city crime data
from Police Department. The results of this data mining could
potentially be used to lessen and even prevent crime for the
forth coming years.Dr. S. Santhosh Baboo and Malathi. A [7]
research work focused on developing a crime analysis tool for
Indian scenario using different data mining techniques that
can help law enforcement department to efficiently handle
crime investigation. The proposed tool enables agencies to
easily and economically clean, characterize and analyze crime
data to identify actionable patterns and trends .Kadhim
B. Swadi Al-Janabi [8] presents a proposed framework for the
crime and criminal data analysis and detection using Decision
tree Algorithms for data classification and Simple K Means
algorithm for data clustering. The paper tends to help
specialists in discovering patterns and trends, making
forecasts, finding relationships and possible explanations,
mapping criminal networks and identifying possible suspects.
Aravindan Mahendiran et al. [9] apply myriad of tools on
crime data sets to mine for information that is hidden from
human perception. With the help of state of the art
visualization techniques we present the patterns discovered
through our algorithms in a neat and intuitive way that enables
law enforcement departments to channelize their resources
accordingly. Sutapat Thiprungsri[10] examine the possibility
of using clustering technology for auditing. Automating fraud
filtering can be of great value to continuous audits. The
objective of their study is to examine the use of cluster
analysis as an alternative and innovative anomaly detection
technique in the wire transfer system. K. Zakir Hussain et al.
[11] tried try to capture years of human experience into
computer models via data mining and by designing a
simulation model.
3. PROPOSED SYSTEM
ARCHITECTURE
After literature review there is need to used an open source
data mining tool which can be implemented easily and
analysis can be done easily. So here crime analysis is done on
crime dataset by applying k means clustering algorithm using
rapid miner tool.
The procedure is given below:
1. First we take crime dataset
2. Filter dataset according to requirement and create new
dataset which has attribute according to analysis to be
done
3. Open rapid miner tool and read excel file of crime
dataset and apply “Replace Missing value operator” on it
and execute operation
4. Perform “Normalize operator” on resultant dataset and
execute operation
5. Perform k means clustering on resultant dataset formed
after normalization and execute operation
6. From plot view of result plot data between crimes and
get required cluster
7. Analysis can be done on cluster formed.
Fig 1: Flow chart of crime analysis
4. EXPERIMENTAL SETUP AND
RESULTS
4.1 Approach Used
4.1.1 k-means algorithm
K-means clustering is one of the method of cluster
analysis which aims to partition n observations into k clusters
in which each observation belongs to the cluster with the
nearest mean.
Process
1. Initially, the number of clusters must be known let it be k
2. The initial step is the choose a set of K instances as
centres of the clusters.
3. Next, the algorithm considers each instance and assigns
it to the cluster which is closest.
4. The cluster centroids are recalculated either after whole
cycle of re-assignment or each instance assignment.
5. This process is iterated.
K means algorithm complexity is O(tkn), where n is
instances, c is clusters, and t is iterations and relatively
efficient . It often terminates at a local optimum. Its
disadvantage is applicable only when mean is defined and
need to specify c, the number of clusters, in advance. It unable
to handle noisy data and outliers and not suitable to discover
clusters with non-convex shapes.
Take crime dataset
Filter dataset according to
requirement
Open Rapid miner tool and
read excel file of crime dataset
Apply Replace Missing Value
operator and execute
Perform k means clustering on
resultant dataset and execute
Perform Normalization operator
on resultant dataset and execute
Perform plot view and get cluster
Perform crime analysis on cluster
formed

International Journal of Computer Applications (0975 – 8887)
Volume 83 – No4, December 2013
3
4.2 Dataset Used
Crime dataset used for crime analysis is an offences recorded
by the police in England and Wales by offence and police
force area from 1990 to 2011-12 [12].In Table 1 sample crime
dataset is shown.
Table 1. Crime dataset
Year
Homicide
Attempted
murder
Child
destruction
Causing
death by
careless
driving
1990
10
19
0
7
1990
6
10
0
5
1990
6
8
0
9
1990
6
2
0
15
1990
10
5
0
1
4.3 Tool Used
Many open source data mining suites are available such as R,
Tanagra, Weka , KNIME, Orange, Rapid miner. Here we are
performing crime analysis using Rapid miner tool because of
following reason:
1. It is solid and complete package with Flexible/affordable
support options.
2. Enterprise-ready performance and scalability for big
data analytics Innovative analyst support
3. We can program by piping components together in a
graphic ETL work flows.
Also it has good features that if you set up an illegal work
flows Rapid Miner suggest Quick Fixes to make it legal.
4.4. K means cluster analysis
This involves tracking crime rate changes from one year to the
next and used data mining to project those changes into the
future. Here we consider homicide crime and plot it with year
and analysis variation in graph on cluster formed.
1. Homicide
Cluster 0
Fig 2: Homicide is minimum in 2004 and maximum and
same in 2000 & 2008
From Fig 2 it can be seen that in year 2004 number of
homicide crime committed is minimum as compared to in
year 2008 where maximum number of homicide crime
committed.
Cluster 1
Fig 3: Homicide is minimum in 2008 and maximum in
1990 & 2004.
From Fig 3 it can be seen that in year 2008 number of
homicide crime committed is minimum as compared to in
year 1990 and 2000 where maximum number of homicide
crime committed.
Cluster 2
Fig 4: Homicide is minimum in 1992 and maximum in
2002
From Fig 4 it can be seen that in year 1992 number of
homicide crime committed is minimum as compared to in
year 2002 where maximum number of homicide crime
committed.
Cluster 3
Fig 5: Homicide is minimum in 2011 and maximum in
2003
0
5
10
15
1990
1993
1996
1999
2002
2005
2008
2011
no. of
crime
year
homicide
homicide
0
20
40
60
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
no. of
crime
year
homicide
homicide
0
200
400
600
1990
1993
1996
1999
2002
2005
2008
2011
no. of
crime
year
homicide
homicide
0
200
400
600
1990
1993
1996
1999
2002
2005
2008
2011
no. of
crime
year
homicide
homicide

International Journal of Computer Applications (0975 – 8887)
Volume 83 – No4, December 2013
4
From Fig 5 it can be seen that in year 2011 number of
homicide crime committed is minimum as compared to in
year 2003 where maximum number of homicide crime
committed.
Cluster 4
Fig 6: Homicide is minimum in 1990 & 1993 and
maximum in 2007
From Fig 6 it can be seen that in year 1990 and 1993 number
of homicide crime committed is minimum as compared to
year 2007 where maximum number of homicide crime
committed.
5. CONCLUSION
This project focuses on crime analysis by implementing
clustering algorithm on crime dataset using rapid miner tool
and here we do crime analysis by considering crime homicide
and plotting it with respect to year and got into conclusion
that homicide is decreasing from 1990 to 2011 .From the
clustered results it is easy to identify crime trend over years
and can be used to design precaution methods for future.
6. FUTURE SCOPE
From the encouraging results, we believe that crime data
mining has a promising future for increasin the effectiveness
and efficiency of criminal and intelligence analysis. Visual
and intuitive criminal and intelligence investigation
techniques can be developed for crime pattern. As we have
applied clustering technique of data mining for crime analysis
we can also perform other techniques of data mining such as
classification. Also we can perform analysis on various
dataset such as enterprise survey dataset, poverty dataset, aid
effectiveness dataset, etc.
7. REFERENCES
[1] De Bruin ,J.S.,Cocx,T.K,Kosters,W.A.,Laros,J. and
Kok,J.N(2006) Data mining approaches to criminal
carrer analysis ,”in Proceedings of the Sixth International
Conference on Data Mining (ICDM”06) ,Pp. 171-177
[2] Manish Gupta1*, B.Chandra1 and M. P. Gupta1,2007
Crime Data Mining for Indian Police Information System
[3] Nazlena Mohamad Ali1, Masnizah Mohd2, Hyowon
Lee3, Alan F. Smeaton3, Fabio Crestani4 and Shahrul
Azman Mohd Noah2 ,2010 Visual Interactive Malaysia
Crime News Retrieval System
[4] Sutapat Thirprungsri Rutgers University .USA ,2011
Cluster Analysis of Anomaly Detection in Accounting
Data : An Audit Approach 1
[5] A.Malathi ,Dr.S.Santhosh Baboo. D.G. Vaishnav
College,Chennai ,2011 Algorithmic Crime Prediction
Model Based on the Analysis of Crime Clusters.
[6] Malathi.A 1 ,Dr.S.Santhosh Baboo 2 and Anbarasi . A 31
Assistant professor ,Department of Computer Science
,Govt Arts College ,Coimbatore , India . 2 Readers ,
Department of Computer science , D.G. Vaishnav Collge
,Chennai , India , 2011 An intelligent Analysis of a city
Crime Data Using Data Mining
[7] Malathi , A; Santhosh Baboo , S, 2011 An Enhanced
Algorithm to Predict a Future Crime using Data Mining
[8] Kadhim B.Swadi al-Janabi . Department of Computer
Science . Faculty of Mathematics and Computer Science
.University of Kufa/Iraq , 2011 A Proposed Framework
for Analyzing Crime DataSet using Decision Tree and
Simple K-means Mining Algorithms.
[9] Aravindan Mahendiran, Michael Shuffett, Sathappan
Muthiah, Rimy Malla, Gaoqiang Zhang,2011 Forecasting
Crime Incidents using Cluster Analysis and Bayesian
Belief Networks
[10] Sutapat Thiprungsri,2012 Cluster Analysis for Anomaly
Detection in Accounting Data : An Audit Approach1
[11] K. Zakir Hussain, M. Durairaj and G. Rabia Jahani
Farzana ,2012 Application of Data Mining Techniques
for Analyzing Violent Criminal Behavior by Simulation
Model
[12] https://www.gov.uk/government/publications/offences-
recorded-by-the-police-in-england-and-wales-by-
offence-and-police-force-area-1990-to-2011-12
0
200
400
1990
1993
1996
1999
2002
2005
2008
2011
no. of
crime
year
homicide
homicide
IJCATM : www.ijcaonline.org