CRIME ANALYSIS AND PREDICTION USING MACHINE LEARNING

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 09 | Sep 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 916
CRIME ANALYSIS AND PREDICTION USING MACHINE LEARNING
Roopa1, Prof. Thouseef Ulla Khan2
1PG Scholar (MCA), Dept of MCA, Vidya Vikas Institute of Engineering And Technology, Mysore ,Karnataka, India
2Assistant Professor, Dept of MCA, Vidya Vikas Institute of Engineering And Technology, Mysore ,Karnataka, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract – Crime is one of our society's most serious and
pervasive problems, and preventing it is a critical duty. This
necessitates keeping note of all offences and creating a
database for future reference. The present issue is keeping a
reliable crime record and analysing this data to aid in the
prediction and resolution of futurecrimes. Theobjectiveof this
paper is to analyze dataset, which consist of numerous crimes
and predicting the type of crime, which may happen in future
depending upon various conditions. In this project, we will be
using the technique of machine learning and data science for
crime prediction of Indian crime data set. Crime analysis and
prediction is a methodical way to spotting crime. This
algorithm can anticipate and depict crime-prone areas. Using
the notion of machine learning, we may extract previously
unknown, meaningful information from unstructured data.
The extraction of new informationisanticipatedusingcurrent
datasets. Crime is a perilousand widespreadsocietalissuethat
affects people all around the world. Crime has an impact on
people's quality of life, economic prosperity, and the nation's
reputation. To safeguard their communities from crime,
modern technology and novel techniques to enhancing crime
analytics are required. We present a system that can analyse,
identify, and forecast various crime probabilities in a given
location. This study describes many sorts of criminal analysis
and crime prediction using machine learning approaches.
Key Words: Decision trees, linear regression, and k-means
clustering
1. INTRODUCTION
The crime data rate is growing on a daily basis because
current technology and high-tech ways assist criminals in
carrying out their illicit actions. Burglary, arson, and other
crimes, according to the Crime Record Bureau have
escalated, as have crimes such as murder, rape, abuse, gang
rap, and so on. Data on crime will be gathered from
numerous blogs, news sites, and websites. The massive
amount of data is utilized to create a record. A database of
crime reports. The knowledge gained via data mining
techniques will be useful in lowering crime by making it
easier to discover the perpetrators and the regions most
affected by crime.
When applied to a crime dataset, data mining techniques
produce good results. The information generated from data
mining techniques can assist the police department. The
discovery of criminal "hot spots," which show regionswitha
high concentration of crime, hasbeenprovenvaluablebythe
police. Data mining approaches can yield significantfindings
from crime report databases. Crime analysis is the first
phase in the study of crime. Criminal analysis is the
exploration, interrelationship,anddetectionofrelationships
between numerous crimesandcrimevariables.Thisanalysis
aids in the creation of statistics, queries, and maps on
demand. It also aids in determining whether a crime has
occurred in a certain recognized location.
1.1 Objectives
The prediction using data mining techniques that is
prediction rules. Frequent patterns are extracted based on
the criteria’s like crime type. Prediction is done basedonthe
previous year datasets. The prediction report consists of all
the datasets from the year 2012-2020.the year wise
comparison is shown based on the state wise datasets. The
clustering algorithm can be perform basedoneverydatasets
based on each year wise comparison is made.
1.2 Scope
The primary goals of crime evaluations are as follows: 1.
Identifying crime tendencies by study of existing crimesand
criminal information 2. Using geographic distribution to
forecast crime of available information as well as prediction
ofcrime total utilizing various datamining techniques 3.
Criminal detection
2. Existing System:
This algorithm can forecast high-risk areas for crime and
show crime-prone areas. Using the concept of data mining,
we may draw previously undiscovered, pertinent
information from unstructured data.
Disadvantages:
 They exhibited lower prediction accuracy using
this technique.
 The outcomes of this approach are not ideal.

3. Planned system:
The system under consideration is a web-based application.
An advanced criminal mechanism of detection whose main
goal is to forecast crimes and their tendencies.Theproposed
system employs a data mining approach known as
"Prediction Rules" for crime pattern detection, as well as
automation for early crime pattern prediction, which helps
to avert crimes. Predicts crime trends based on past crime
information, date, and location.
Advantages:
 Conducting criminal analysis and identifying trends
in crime.
 Disseminate knowledge to help with the creation of
crime reduction and preventive measures.
 Recognize and examine recurring criminal trends to
prevent similar incidents from happening again.
 To create a data-cleaning algorithm that purges the
crime dataset of unnecessary information so that it
may be explored.
4. System Design
Designing a machine learning system's software
architecture, infrastructure, algorithms, and data to meet
specific needs is known as machine learning systems
architecture.By outlining the intricacies of how the
programme should be created, the software design will be
used to assist in the development of software for web apps.
Fig: 4.2.1Architecture Design
5. Detailed design
By outlining the specifics of how the application should be
constructed, the software design will be utilised to assist in
the software development of an android application. Use
case models, sequence diagrams, and other supplementary
requirement data are included in the software design
specifications, which are narrative and graphical
documentation of the software design
5.1 Diagram of Use Cases:
An example of a behavioural diagram in the Unified
Modelling Language (UML) is a use case diagram, which is
based on and defined by use case studies. Its purpose is to
present a graphical depiction of a system's functioning in
terms of actors, their objectives (represented as use cases),
and any connections between those use cases. The main
objective of a use case diagram is to show which system
actions are taken for a particular actor. You may display the
parts that each player in the system plays
6. Implementation
To forecast the future values of the dependent variable, a
regression algorithm is created to identify the past link
between an independent and a dependent variable.
Regression uses the historical relationship between
variables to forecast how they will behave going forward.

Investigative work on the sites of high and low frequency
crimes was done using the K-means clustering technique.
6.1 Algorithm Implementation
6.1.1Regression Algorithm
A regression algorithm is designed to find the historical
relationship between an independent and a dependent
variable to predict the future values of the dependent
variable. In order to forecast future behaviour, a regression
models the historical connection between variables. The
Algorithm uses thelinearregressiontechniquesbasedon the
data set collected for the project. The linear regression
technique helps in predicting the future behavior of road
CRIME with help of the statistical methods. The algorithm
find the mean and variance value ofthedependentvariables,
and apply the formula Y=b0+b1*x to predict the future
behavior.
6.1.2 K-means Clustering
K-means clustering algorithm was used to investigate the
high and low-frequency CRIME locations. The algorithm
follows a simple and easy way to classify a given data set
through a certain number of clusters (assume k clusters)
fixed a priori. M These centroids should be placed in a
cunning way because of different location causes different
result.
6.1.3 Decision Tree:
For both prediction and classification, a decision tree is
employed. A function, which is intervals formed by splits
on the individuals' attributes value, may be trained for
classification purposes.
X
(year)
Y(value) A1=(x-
mean
of x)
B1=(y—
mean of
y)
A1
*B1
(A1)2 (b1)2
2008 3496 -4 -329 1316 16 108241
2010 3500 -3 -325 975 9 105625
2010 3987 -2 162 -324 4 26244
2011 2987 -1 -838 838 1 702244
2012 3019 0 -806 0 0 649636
2013 3999 1 174 174 1 30276
2014 4015 2 190 380 4 36100
2015 4786 3 961 2883 9 923521
2016 4018 4 193 772 16 37249
2021 4445 5 620 3100 25 384400
Fig 6.4.10To display the crime rates year in India
7. Testing
This chapter gives the various test cases performed tocheck
for the effective execution of the venture. Testing is a
procedure of cross verificationofthedesignedsystemmodel
under active state and various inputs. There are several
ways to carry out this approach. The main objective of
software development life cycle is to producea productwith
no errors or very few errors. In the processes of achieving
hassle free software we plan testing and test cases. Software
testing is done for the success of the application. The testing
is done mainly to check whether the product meet the
requirement of the user properly. It is usedtocheck the bugs
and errors in the system or to find out the defects of the
system.
7.1 Test causes with positive scenarios:
TC
No
Positive
scenario
Required
Input
Expected
output
Actual output Test
Result
1 Enter
Prediction
values
Enter a valid
values
Should
predicted
successfully
predicted
successfully
Pass
2 Enter
clustering
values
State,year,
type
Should
cluster
successfully
cluster
successfully
Pass
3 Enter
Prediction
values
Enter a valid
values
Should
predicted
successfully
Database
error
fail
4 Enter
clustering
values
State,year,
type
Should
cluster
successfully
Database
error
Fail
Conclusion
Due to a variety of variables, including a growth in poverty,
unemployment, corruption, etc., crime rates in India are
rising daily. The suggested paradigm is extremely beneficial
to both the investigating authorities and the taking the
required actions as a police officer to lower crime. The
initiative aids in the examination of these crimes by using

various interactive visualisationtechniques,crimenetworks
Future development of this project will focus on teaching
bots to identify crime hotspots using machine learning
acquiring skills. Giventhatmachinelearninganddata mining
are comparable, an advanced machine learning concept can
be applied to improve prediction. It is possible to increase
the data's dependability,correctness,andprivacyprediction.
REFERENCES:
[1] Ginger Saltos and Mihaela Coacea, 2017 International
Journal of Information Technology and Decision Making, An
Exploration of Crime Prediction Using Data Mining on Open
Data.
[2] Shiju Sathyadevan, Devan M.S, Surya Gangadharan.S,
Crime Analysis and Prediction Using First International
Conference on networks & soft computing (IEEE) 2014.
[3] Khushabu A.Bokde, Tisksha P.Kakade, Dnyaneshwari S.
Tumasare, Chetan G.Wadhai B.E Student, Crime Detection
Techniques Using and K-Means, International Journal of
Engineering Research & technology (IJERT) ,2018
[4] Crime Pattern Analysis, Visualization And Prediction
Using Data Mining, Indian Journal of Computer Science and
Engineering(IJCSE),Tushar Sonawanev,ShirinShaikh,Rahul
Shinde, and Asif Sayyad, 2015.
[5] Raj Kumar and Sakkarai Pandi, "Crime Analysis and
Prediction Using Data Mining Techniques," International
Journal of Recent Trends in Engineering & Research, 2019.
[6] Sarpreet kaur, Dr. Williamjeet Singh, Systematic review
of machine learning using python, International Journal of
Advanced Research in computer science, 2015.
[7] Kalyani Kadam and AyisheshimAlmaw,"SurveyPaper on
Crime Prediction Using Ensemble Approach," International
Journal of Pure and Applied Mathematics, 2018.
[8] Review on Crime Analysis and Prediction Using Data
Mining Techniques, International Journal of Innovative
Research in Science Engineering and Technology, 2018, by
Dr. M. Sreedevi, A. Hardhat Vardhan Reddy, and Ch. Venkata
Sai Krishna Reddy
[9] International journal of engineering, Science and
mathematics, 2017. K.S.N. Murthy, A.V.S. Pavan Kumar, and
Gangu Dharmaraju

CRIME ANALYSIS AND PREDICTION USING MACHINE LEARNING

More Related Content

What's hot (20)

Similar to CRIME ANALYSIS AND PREDICTION USING MACHINE LEARNING (20)

More from IRJET Journal (20)

Recently uploaded (20)

CRIME ANALYSIS AND PREDICTION USING MACHINE LEARNING