International Journal of Computer Applications (0975 – 8887)

Volume 83 – No4, December 2013

Crime Analysis using K-Means Clustering

Jyoti Agarwal Renuka Nagpal Rajni Sehgal

Mtech CSE Assistant Professor Assistant Professor
Amity University,Noida Amity University ,Noida Amity University,Noida

ABSTRACT trend correlations to assist personnel in planning the

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

1
 International Journal of Computer Applications (0975 – 8887)
Volume 83 – No4, December 2013

filtering during an audit. He used cluster analysis to help

auditors focus their efforts when evaluating group life
Take crime dataset
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
Filter dataset according to
clustering/classify based model to anticipate crime trends. The
data mining techniques are used to analyze the city crime data requirement
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] Open Rapid miner tool and
research work focused on developing a crime analysis tool for read excel file of crime dataset
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
Apply Replace Missing Value
data to identify actionable patterns and trends .Kadhim
B. Swadi Al-Janabi [8] presents a proposed framework for the operator and execute
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 Perform Normalization operator
specialists in discovering patterns and trends, making on resultant dataset and execute
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 Perform k means clustering on
human perception. With the help of state of the art resultant dataset and execute
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 Perform plot view and get cluster
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 Perform crime analysis on cluster
technique in the wire transfer system. K. Zakir Hussain et al. formed
[11] tried try to capture years of human experience into
computer models via data mining and by designing a
simulation model. Fig 1: Flow chart of crime analysis

3. PROPOSED SYSTEM 4. EXPERIMENTAL SETUP AND

ARCHITECTURE RESULTS
After literature review there is need to used an open source 4.1 Approach Used
data mining tool which can be implemented easily and 4.1.1 k-means algorithm
analysis can be done easily. So here crime analysis is done on K-means clustering is one of the method of cluster
crime dataset by applying k means clustering algorithm using analysis which aims to partition n observations into k clusters
rapid miner tool. in which each observation belongs to the cluster with the
nearest mean.
The procedure is given below:
Process
1. First we take crime dataset
2. Filter dataset according to requirement and create new 1. Initially, the number of clusters must be known let it be k
dataset which has attribute according to analysis to be 2. The initial step is the choose a set of K instances as
done centres of the clusters.
3. Open rapid miner tool and read excel file of crime 3. Next, the algorithm considers each instance and assigns
dataset and apply “Replace Missing value operator” on it it to the cluster which is closest.
and execute operation 4. The cluster centroids are recalculated either after whole
4. Perform “Normalize operator” on resultant dataset and cycle of re-assignment or each instance assignment.
execute operation 5. This process is iterated.
5. Perform k means clustering on resultant dataset formed K means algorithm complexity is O(tkn), where n is
after normalization and execute operation instances, c is clusters, and t is iterations and relatively
6. From plot view of result plot data between crimes and efficient . It often terminates at a local optimum. Its
get required cluster disadvantage is applicable only when mean is defined and
7. Analysis can be done on cluster formed. 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.

2
 International Journal of Computer Applications (0975 – 8887)
Volume 83 – No4, December 2013

4.2 Dataset Used

Crime dataset used for crime analysis is an offences recorded From Fig 2 it can be seen that in year 2004 number of
by the police in England and Wales by offence and police homicide crime committed is minimum as compared to in
force area from 1990 to 2011-12 [12].In Table 1 sample crime year 2008 where maximum number of homicide crime
dataset is shown. committed.
Cluster 1
Table 1. Crime dataset

Year Homicide Attempted Child Causing homicide

murder destruction death by
careless 60
driving
no. of 40
crime 20
1990 10 19 0 7 0 homicide

1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
1990 6 10 0 5 year

1990 6 8 0 9
Fig 3: Homicide is minimum in 2008 and maximum in
1990 & 2004.
1990 6 2 0 15
From Fig 3 it can be seen that in year 2008 number of
1990 10 5 0 1
homicide crime committed is minimum as compared to in
year 1990 and 2000 where maximum number of homicide
crime committed.
4.3 Tool Used Cluster 2
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 homicide
following reason:
600
1. It is solid and complete package with Flexible/affordable no. of 400
support options. crime 200
2. Enterprise-ready performance and scalability for big 0 homicide
data analytics Innovative analyst support
1993
1990

1996
1999
2002
2005
2008
2011
3. We can program by piping components together in a
graphic ETL work flows.
year
Also it has good features that if you set up an illegal work
flows Rapid Miner suggest Quick Fixes to make it legal.
Fig 4: Homicide is minimum in 1992 and maximum in
4.4. K means cluster analysis 2002
This involves tracking crime rate changes from one year to the
next and used data mining to project those changes into the From Fig 4 it can be seen that in year 1992 number of
future. Here we consider homicide crime and plot it with year homicide crime committed is minimum as compared to in
and analysis variation in graph on cluster formed. year 2002 where maximum number of homicide crime
1. Homicide committed.
Cluster 0
Cluster 3

homicide
homicide
15
600
no. of 10
no. of 400
crime 5
crime 200
0 homicide 0 homicide
1999
1990
1993
1996

2002
2005
2008
2011

1993
1990

1996
1999
2002
2005
2008
2011

year year

Fig 2: Homicide is minimum in 2004 and maximum and Fig 5: Homicide is minimum in 2011 and maximum in
same in 2000 & 2008 2003

3
 International Journal of Computer Applications (0975 – 8887)
Volume 83 – No4, December 2013

From Fig 5 it can be seen that in year 2011 number of 7. REFERENCES

homicide crime committed is minimum as compared to in [1] De Bruin ,J.S.,Cocx,T.K,Kosters,W.A.,Laros,J. and
year 2003 where maximum number of homicide crime Kok,J.N(2006) Data mining approaches to criminal
committed. carrer analysis ,”in Proceedings of the Sixth International
Conference on Data Mining (ICDM”06) ,Pp. 171-177
Cluster 4
[2] Manish Gupta1*, B.Chandra1 and M. P. Gupta1,2007
Crime Data Mining for Indian Police Information System
homicide
[3] Nazlena Mohamad Ali1, Masnizah Mohd2, Hyowon
400 Lee3, Alan F. Smeaton3, Fabio Crestani4 and Shahrul
Azman Mohd Noah2 ,2010 Visual Interactive Malaysia
no. of
200 Crime News Retrieval System
crime
0 homicide [4] Sutapat Thirprungsri Rutgers University .USA ,2011
2008
1990
1993
1996
1999
2002
2005

2011
Cluster Analysis of Anomaly Detection in Accounting
Data : An Audit Approach 1
year [5] A.Malathi ,Dr.S.Santhosh Baboo. D.G. Vaishnav
College,Chennai ,2011 Algorithmic Crime Prediction
Model Based on the Analysis of Crime Clusters.
Fig 6: Homicide is minimum in 1990 & 1993 and
maximum in 2007 [6] Malathi.A 1 ,Dr.S.Santhosh Baboo 2 and Anbarasi . A 31
Assistant professor ,Department of Computer Science
From Fig 6 it can be seen that in year 1990 and 1993 number ,Govt Arts College ,Coimbatore , India . 2 Readers ,
of homicide crime committed is minimum as compared to Department of Computer science , D.G. Vaishnav Collge
year 2007 where maximum number of homicide crime ,Chennai , India , 2011 An intelligent Analysis of a city
committed. Crime Data Using Data Mining
[7] Malathi , A; Santhosh Baboo , S, 2011 An Enhanced
5. CONCLUSION Algorithm to Predict a Future Crime using Data Mining
This project focuses on crime analysis by implementing
clustering algorithm on crime dataset using rapid miner tool [8] Kadhim B.Swadi al-Janabi . Department of Computer
and here we do crime analysis by considering crime homicide Science . Faculty of Mathematics and Computer Science
and plotting it with respect to year and got into conclusion .University of Kufa/Iraq , 2011 A Proposed Framework
that homicide is decreasing from 1990 to 2011 .From the for Analyzing Crime DataSet using Decision Tree and
clustered results it is easy to identify crime trend over years Simple K-means Mining Algorithms.
and can be used to design precaution methods for future. [9] Aravindan Mahendiran, Michael Shuffett, Sathappan
Muthiah, Rimy Malla, Gaoqiang Zhang,2011 Forecasting
Crime Incidents using Cluster Analysis and Bayesian
6. FUTURE SCOPE Belief Networks
From the encouraging results, we believe that crime data
mining has a promising future for increasin the effectiveness [10] Sutapat Thiprungsri,2012 Cluster Analysis for Anomaly
and efficiency of criminal and intelligence analysis. Visual Detection in Accounting Data : An Audit Approach1
and intuitive criminal and intelligence investigation
techniques can be developed for crime pattern. As we have [11] K. Zakir Hussain, M. Durairaj and G. Rabia Jahani
applied clustering technique of data mining for crime analysis Farzana ,2012 Application of Data Mining Techniques
we can also perform other techniques of data mining such as for Analyzing Violent Criminal Behavior by Simulation
classification. Also we can perform analysis on various Model
dataset such as enterprise survey dataset, poverty dataset, aid [12] https://www.gov.uk/government/publications/offences-
effectiveness dataset, etc. recorded-by-the-police-in-england-and-wales-by-
offence-and-police-force-area-1990-to-2011-12

IJCATM : www.ijcaonline.org 4