International Journal of Scientific Research in Science, Engineering and Technology

Print ISSN - 2395-1990 Available Online at : [Link]

Online ISSN : 2394-4099 doi : [Link]

Crime Prediction Using Machine Learning and Deep Learning

P. Karthik*1 , P. Jayanth2, K. Tharun Nayak3 , K. Anil Kumar4
*1 UG Student, Department of Computer Science and Engineering (Internet of Things), Guru Nanak Institutions
Technical Campus, Hyderabad, Telangana, India
4Assistant Professor, Department of Computer Science and Engineering (Internet of Things), Guru Nanak
Institutions Technical Campus, Hyderabad, Telangana, India

ARTICLEINFO ABSTRACT

The utilization of machine learning and deep learning methods for crime
Article History:
prediction has become a focal point for researchers, aiming to decipher the
Accepted: 01 May 2024 complex patterns and occurrences of crime. This review scrutinizes an
Published: 05 May 2024 extensive collection of over 150 scholarly articles to delve into the
assortment of machine learning and deep learning techniques employed in
forecasting criminal behaviour. It grants access to the datasets leveraged
Publication Issue : by researchers for crime forecasting and delves into the key methodologies
Volume 11, Issue 3 utilized in these predictive algorithms. The study sheds light on the
May-June-2024 various trends and elements associated with criminal behaviour and
Page Number : underscores the existing deficiencies and prospective avenues for
08-15 advancing crime prediction precision. This thorough examination of the
current research on crime forecasting through machine learning and deep
learning serves as an essential resource for scholars in the domain. A more
profound comprehension of these predictive methods will empower law
enforcement to devise more effective prevention and response strategies
against crime.
Keywords : Machine Learning , Deep Learning , Crime Prediction ,
Research Review , Algorithm Application, Dataset Analysis , Trend
Identification , Criminal Activity Factors , Predictive Accuracy , Future
Directions, Law Enforcement Strategies

I. INTRODUCTION accuracy, supporting safer communities and smarter

law enforcement. Understanding the factors behind
Crime prediction integrates data science and machine crime helps address the challenges of urbanization and
learning to anticipate crimes, aiding proactive policing. crime prevention.
It uses historical data and algorithms to identify crime
patterns and inform resource allocation. Advances in
machine learning have enhanced crime prediction

Copyright © 2024 The Author(s): This is an open-access article distributed under the terms of the Creative 8
Commons Attribution 4.0 International License (CC BY-NC 4.0)
 P. Karthik et al Int J Sci Res Sci Eng Technol, May-June-2024, 11 (3) : 08-15

II. EXISTING SYSTEM Proposed System Advantages:

1. High Accuracy: The system’s use of Decision Tree
The crime prediction system uses the Random Forest and Bagging Classifiers significantly boosts the
Classifier to forecast and categorize crimes effectively. accuracy of crime predictions.
It stands out for its accuracy, handling complex data, [Link] and Interpretability: The Bagging
and combining multiple decision trees for reliable Classifier enhances robustness against overfitting,
predictions. The dataset contains 15 features, aiding in while the Decision Tree Classifier maintains
comprehensive analysis. Tailored for law enforcement transparency in the predictive process.
and policymakers, the system supports strategic
decision-making and enhances public safety, IV. SYSTEM ARCHITECTURE
demonstrating high accuracy without notable
drawbacks. Our research outlines machine learning-based system
architecture designed to differentiate between fake
Existing System Disadvantages : and real accounts. It starts with a dataset that
1. Computational Intensity: The system’s need for undergoes pre-processing and feature selection to
substantial computational power can be a hurdle in refine the data. This prepared data is then input into
real-time applications or when resources are limited. classifiers like Random Forest and Decision Tree,
2. Lack of Interpretability: The complexity of the which predict the authenticity of an account. The
Random Forest algorithm can make it difficult to results are labelled as either ‘Fake’ or ‘Real’ accounts,
understand and explain the model’s decisions, affecting followed by a performance analysis and graphical
transparency. representation of the outcomes.

III. PROPOSED SYSTEM In essence, the architecture captures the entire process
from raw data to actionable insights. It emphasizes the
The enhanced crime prediction system utilizes Python importance of data preparation and the use of
and integrates Decision Tree and Bagging Classifiers to ensemble learning methods to ensure accurate
improve accuracy in Portland’s crime analysis from predictions. The final step involves analysing the
2015 to 2023. It maintains high accuracy, processes a model’s performance, likely to assess its effectiveness
vast dataset with over half a million data points, and and accuracy in classifying accounts. This systematic
classifies 20 different crime types. Features like address, approach is crucial for maintaining the integrity of
case number, and offense details provide rich input for digital platforms by identifying and filtering out
the models. fraudulent accounts.

This system is designed for adaptability and scalability,

with user-friendly interfaces for stakeholders. It
includes mechanisms for continuous model updates
and hyperparameter tuning to adapt to changing crime Figure 1 : System architecture
patterns. The goal is to refine crime prediction
techniques, optimize accuracy, and overcome previous V. LITERATURE SURVEY
limitations, making it a robust tool for law
enforcement and policymakers. This paper explores the use of machine learning (ML)
and computer vision to predict and prevent crime.

International Journal of Scientific Research in Science, Engineering and Technology | [Link] | Vol 11 | Issue 3 9
 P. Karthik et al Int J Sci Res Sci Eng Technol, May-June-2024, 11 (3) : 08-15

With crime rates rising, traditional methods fall short, Crimes pose a significant challenge to justice and
prompting the need for innovative solutions. Our require effective control measures. This study
research demonstrates how ML and computer vision leverages various machine learning algorithms,
can transform law enforcement, offering faster, more including logistic regression, SVM, Naïve Bayes, KNN,
accurate crime detection and prevention. We present decision tree, MLP, random forest, XGBoost, and time
case studies that validate the effectiveness of these series analysis with LSTM and ARIMA models, to
technologies, suggesting a significant shift towards enhance crime prediction and forecasting. The LSTM
more advanced crime-fighting strategies. model showed promising results in terms of RMSE and
MAE. Data analysis indicates a potential decrease in
This project evaluates the use of neural networks in Chicago’s crime rate and a minor rise in Los Angeles’s,
machine learning to predict criminal behaviour from with February having fewer incidents. Predictions
arrest records, addressing imbalanced data with data suggest a moderate future increase in Chicago’s crime
augmentation and a weighted loss function. Focusing rate with a possible subsequent decline, while Los
on deep fully connected neural networks, suitable for Angeles is expected to see a significant decrease as per
complex feature relationships and limited domain the ARIMA model. These insights help in early crime
knowledge, the study underscores the potential of detection, identifying high-risk areas, and informing
machine learning in crime classification and suggests police strategies for improved safety.
further exploration of these techniques for enhanced
predictive accuracy. Crime significantly disrupts societal norms and
hampers the social and economic stability of nations, a
Ensemble learning is a collective approach that situation that Bangladesh also faces. This study
combines the outputs of various classifiers to generate concentrates on forecasting crime-prone areas, which
new predictions. Studies indicate that these ensemble can be instrumental in anticipating criminal incidents
models tend to be more dependable than single across various cities within specific timeframes. With
classifiers. Despite the availability of numerous insights into upcoming trends, law enforcement and
ensemble techniques, selecting the right one for a detectives can implement preventive measures to
specific dataset remains challenging. In the context of decrease crime rates. The Random Forest algorithm is
crime prediction in India, this complexity is employed in this research to identify districts with
heightened by the dynamic nature of criminal varying levels of criminal activity.
activities. Our research introduces a method called
assemble-stacking based crime prediction method [Link] ALGORITHM
(SBCPM), utilizing SVM algorithms to enhance crime
prediction accuracy using MATLAB. This method is Random Forest Classifier : The Random Forest
compared with other models like J48, SMO, Naïve Classifier is a sophisticated machine learning algorithm
Bayes bagging, and Random Forest, showing that that belongs to the ensemble learning family. It
ensemble models can sometimes surpass others, constructs multiple decision trees during the training
especially in terms of correlation and error metrics. phase and outputs the class that is the mode of the
Achieving a 99.5% accuracy rate in tests, our method classes for classification tasks, or the mean prediction
outperforms previous studies and aligns well with for regression tasks. This method is known for its high
criminological theories, proving effective for accuracy, ability to handle large datasets with higher
predicting crimes and suggesting that ensemble models dimensionality, and robustness to overfitting. Random
may offer superior accuracy over individual classifiers. Forest works by creating a ‘forest’ of decision trees

International Journal of Scientific Research in Science, Engineering and Technology | [Link] | Vol 11 | Issue 3 10
 P. Karthik et al Int J Sci Res Sci Eng Technol, May-June-2024, 11 (3) : 08-15

from randomly selected subsets of the training set and

then averaging the predictions to improve the overall
result .

VII. PROPOSED ALGORITHM

The Decision Tree Classifier is a type of supervised

learning algorithm that is used for classification tasks. Figure 2: Dataset Upload Page
It works by creating a model that predicts the value of
a target variable by learning simple decision rules
inferred from the data features. It’s like a flowchart
where each internal node represents a test on an
attribute, each branch represents the outcome of the
test, and each leaf node represents a class label . On the
other hand, the Bagging Classifier, also known as
Bootstrap Aggregating, is an ensemble meta-estimator
Figure 3: Model Running
that fits multiple base classifiers (like Decision Trees)
on random subsets of the original dataset. It then
aggregates their individual predictions to form a final
prediction, which helps to improve the stability and
accuracy of machine learning algorithms by reducing
variance and avoiding overfitting .

VIII. RESULTS

User
Input : Enter Login name and Password
Output : If valid user name and password then directly
open the home page otherwise show error message and
redirect to the registration page.

Figure 4: Classification of Crimes based on Dataset

IX. CONCLUSION

To sum up, the “Crime Prediction Using Machine

Figure 1 : Login Page Learning” initiative offers a sophisticated method for
forecasting and sorting a range of criminal offenses in
Portland, Oregon, USA. Utilizing Python’s capabilities
and machine learning techniques like the Decision

International Journal of Scientific Research in Science, Engineering and Technology | [Link] | Vol 11 | Issue 3 11
 P. Karthik et al Int J Sci Res Sci Eng Technol, May-June-2024, 11 (3) : 08-15

Tree and Bagging Classifiers, the project boasts X. REFERENCES

impressive precision, with a 98% accuracy rate for the
training dataset and 95% for the test dataset. [1]. N. Shah, N. Bhagat, and M. Shah, "Crime
The project analyses crime trends from 2015 to 2023 forecasting: A machine learning and computer
using a dataset containing over half a million records, vision approach to crime prediction and
enabling a detailed examination of crime evolution. It prevention," Vis. Comput. Ind., Biomed., Art,
categorizes 20 different types of crimes, ranging from vol. 4, no. 1, pp. 1-14, Apr. 2021.
‘Larceny Offenses’ to ‘Gambling Offenses,’ offering a [2]. S. A. Chun, V. A. Paturu, S. Yuan, R. Pathak, V.
detailed perspective on the area’s crime scene. Atluri, and N. R. Adam, "Crime prediction
The Decision Tree Classifier enhances the system’s model using deep neural networks," in Proc.
transparency, promoting confidence among users, 20th Annu. Int. Conf. Digit. Government Res.,
while the Bagging Classifier increases the system’s Jun. 2019, pp. 512-514.
strength, preventing overfitting and boosting [3]. S. S. Kshatri, D. Singh, B. Narain, S. Bhatia, M. T.
prediction consistency. Quasim, and G. R. Sinha, "An empirical analysis
With a comprehensive set of features like location, case of machine learning algorithms for crime
number, type of crime, neighbourhood, occurrence prediction using stacked generalization: An
date and time, offense category, offense type, and ensemble approach," IEEE Access, vol. 9, pp.
geographical coordinates, the models are well- 67488-67500, 2021.
equipped to deliver well-informed forecasts by [4]. C. Janiesch, P. Zschech, and K. Heinrich,
considering various elements. "Machine learning and deep learning," Electron.
This system not only overcomes the shortcomings of Mark., vol. 31, no. 3, pp. 685-695, Apr. 2021.
previous models but also brings new features such as [5]. W. Safat, S. Asghar, and S. A. Gillani, "Empirical
the ability to adapt to changing crime trends and analysis for crime prediction and forecasting
scalability for various application contexts. The using machine learning and deep learning
addition of user-friendly interfaces and the possibility techniques," IEEE Access, vol. 9, pp. 70080-
of incorporating more data sources improve the 70094, 2021.
system’s usability and the precision of insights for law [6]. S. Kim, P. Joshi, P. S. Kalsi, and P. Taheri, "Crime
enforcement, urban planners, and policy makers. analysis through machine learning," in Proc.
Ultimately, the “Crime Prediction Using Machine IEEE 9th Annu. Inf. Technol., Electron. Mobile
Learning” project emerges as a potent and efficient Commun. Conf. (IEMCON), Nov. 2018, pp. 415-
instrument for improving public safety, optimizing the 420.
distribution of resources, and supporting active [7]. D. M. Raza and D. B. Victor, "Data mining and
decision-making in city settings. The project’s region prediction based on crime using random
achievement in maintaining high accuracy and its forest," in Proc. Int. Conf. Artif. Intell. Smart
comprehensive crime prediction strategy highlight its Syst. (ICAIS), Mar. 2021, pp. 980-987.
significance in bolstering security and shaping well- [8]. L. Elluri, V. Mandalapu, and N. Roy,
informed public policies. "Developing machine learning based predictive
models for smart policing," in Proc. IEEE Int.
Conf. Smart Comput. (SMARTCOMP), Jun.
2019, pp. 198-204.
[9]. A. Meijer and M. Wessels, "Predictive policing:
Review of benefits and drawbacks," Int. J. Public

International Journal of Scientific Research in Science, Engineering and Technology | [Link] | Vol 11 | Issue 3 12
 P. Karthik et al Int J Sci Res Sci Eng Technol, May-June-2024, 11 (3) : 08-15

Admin., vol. 42, no. 12, pp. 1031-1039, Sep. the reduction of crime rates in India," Proc.
2019. Comput. Sci., vol. 172, pp. 122-127, Jan. 2020.
[10]. S. Hossain, A. Abtahee, I. Kashem, M. M. Hoque, [19]. A. Gahalot, S. Dhiman, and L. Chouhan, "Crime
and I. H. Sarker, "Crime prediction using spatio- prediction and analysis," in Proc. 2nd Int. Conf.
temporal data," in Computing Science, Data, Eng. Appl. (IDEA), Feb. 2020, pp. 1-6.
Communication and Security. Gujarat, India: [20]. B. Sivanagaleela and S. Rajesh, "Crime analysis
Springer, 2020, pp. 277-289. and prediction using fuzzy C-means algorithm,"
[11]. M. Saraiva, I. Matijosaitiene, S. Mishra, and A. in Proc. 3rd Int. Conf. Trends Electron.
Amante, "Crime prediction and monitoring in Informat. (ICOEI), Apr. 2019, pp. 595-599.
Porto, Portugal, using machine learning, spatial [21]. A. M. Shermila, A. B. Bellarmine, and N.
and text analytics," ISPRS Int. J. Geo-Inf., vol. Santiago, "Crime data analysis and prediction of
11, no. 7, p. 400, Jul. 2022. perpetrator identity using machine learning
[12]. O. Kounadi, A. Ristea, A. Araujo, and M. Leitner, approach," in Proc. 2nd Int. Conf. Trends
"A systematic review on spatial crime Electron. Informat. (ICOEI), May 2018, pp. 107-
forecasting," Crime Sci., vol. 9, pp. 1-22, Dec. 114.
2020. [22]. C. Catlett, E. Cesario, D. Talia, and A. Vinci, "A
[13]. L. J. Morrisey, "Bibliometric and bibliographic data-driven approach for spatio-temporal crime
analysis in an era of electronic scholarly predictions in smart cities," in Proc. IEEE Int.
communication," in Scholarly Communication Conf. Smart Comput. (SMARTCOMP), Jun.
in Science and Engineering Research in Higher 2018, pp. 17-24.
Education. Evanston, IL, USA: Routledge, 2013, [23]. C. Catlett, E. Cesario, D. Talia, and A. Vinci,
pp. 149-160. "Spatio-temporal crime predictions in smart
[14]. M. Hofmann and A. Chisholm, Text Mining and cities: A data-driven approach and experiments,"
Visualization: Case Studies Using Open-Source Pervas. Mobile Comput., vol. 53, pp. 62-74, Feb.
Tools, vol. 40. Boca Raton, FL, USA: CRC Press, 2019.
2016. [24]. F. Yi, Z. Yu, F. Zhuang, X. Zhang, and H. Xiong,
[15]. P. Tamilarasi and R. U. Rani, "Diagnosis of crime "An integrated model for crime prediction using
rate against women using K-fold cross validation temporal and spatial factors," in Proc. IEEE Int.
through machine learning," in Proc. 4th Int. Conf. Data Mining (ICDM), Nov. 2018, pp. 1386-
Conf. Comput. Methodologies Commun. 1391.
(ICCMC), Mar. 2020, pp. 1034-1038. [25]. S. K. Dash, I. Safro, and R. S. Srinivasamurthy,
[16]. A. Kumar, A. Verma, G. Shinde, Y. Sukhdeve, "Spatio-temporal prediction of crimes using
and N. Lal, "Crime prediction using K-nearest network analytic approach," in Proc. IEEE Int.
neighboring algorithm," in Proc. Int. Conf. Conf. Big Data (Big Data), Dec. 2018, pp. 1912-
Emerg. Trends Inf. Technol. Eng., Feb. 2020, pp. 1917.
1-4. [26]. X. Han, X. Hu, [Link], B. Shen, and [Link], "Risk
[17]. S. Agarwal, L. Yadav, and M. K. Thakur, "Crime prediction of theft crimes in urban communities:
prediction based on statistical models," in Proc. An integrated model of LSTM and ST-GCN,"
11th Int. Conf. Contemp. Comput. (IC), Aug. IEEE Access, vol. 8, pp. 217222-217230, 2020.
2018, pp. 1-3. [27]. Z. Li, C. Huang, L. Xia, Y. Xu, and J. Pei, "Spatial-
[18]. S. R. Bandekar and C. Vijayalakshmi, "Design temporal hypergraph self-supervised learning
and analysis of machine learning algorithms for for crime prediction," in Proc. IEEE 38th Int.

International Journal of Scientific Research in Science, Engineering and Technology | [Link] | Vol 11 | Issue 3 13
 P. Karthik et al Int J Sci Res Sci Eng Technol, May-June-2024, 11 (3) : 08-15

Conf. Data Eng. (ICDE), May 2022, pp. 2984- [35]. A. Baqir, S. U. Rehman, S. Malik, F. U. Mustafa,
2996. and U. Ahmad, "Evaluating the performance of
[28]. N. Tasnim, I. T. Imam, and M. M. A. Hashem, "A hierarchical clustering algorithms to detect
novel multi-module approach to predict crime spatio-temporal crime hot-spots," in Proc. 3rd
based on multivariate spatio-temporal data using Int. Conf. Comput., Math. Eng. Technol.
attention and sequential fusion model," IEEE (iCoMET), Jan. 2020, pp. 1-5.
Access, vol. 10, pp. 48009-48030, 2022. [36]. A. Algefes, N. Aldossari, F. Masmoudi, and E.
[29]. B. Zhou, L. Chen, S. Zhao, S. Li, Z. Zheng, and Kariri, "A text-mining approach for crime tweets
G. Pan, "Unsupervised domain adaptation for in Saudi Arabia: From analysis to prediction," in
crime risk prediction across cities," IEEE Trans. Proc. 7th Int. Conf. Data Sci. Mach. Learn. Appl.
Computat. Social Syst., early access, Sep. 29, (CDMA), Mar. 2022, pp. 109-114.
2022, doi: 10.1109/TCSS.2022.3207987. [37]. S. P. C. W. Sandagiri, B. T. G. S. Kumara, and B.
[30]. U. M. Butt, S. Letchmunan, F. H. Hassan, M. Ali, Kuhaneswaran, "Detecting crime related Twitter
A. Baqir, T. W. Koh, and H. H. R. Sherazi, posts using artificial neural networks based
"Spatio-temporal crime predictions by approach," in Proc. 20th Int. Conf. Adv. ICT
leveraging artificial intelligence for citizens Emerg. Regions (ICTer), Nov. 2020, pp. 5-10.
security in smart cities," IEEE Access, vol. 9, pp. [38]. M. A. Permana, M. I. Thohir, T. Mantoro, and
47516-47529, 2021. M. A. Ayu, "Crime rate detection based on text
[31]. S. Yao, M. Wei, L. Yan, C. Wang, X. Dong, F. mining on social media using logistic regression
Liu, and Y. Xiong, "Prediction of crime hotspots algorithm," in Proc. IEEE 7th Int. Conf.
based on spatial factors of random forest," in Comput., Eng. Design (ICCED), Aug. 2021, pp.
Proc. 15th Int. Conf. Comput. Sci. Educ. 1-6.
(ICCSE), Aug. 2020, pp. 811-815. [39]. X. Zhou, X. Wang, G. Brown, C. Wang, and P.
[32]. M. Sathiyanarayanan, A. K. Junejo, and O. Chin, "Mixed spatiotemporal neural networks
Fadahunsi, "Visual analysis of predictive policing on real-time prediction of crimes," in Proc. 20th
to improve crime investigation," in Proc. Int. IEEE Int. Conf. Mach. Learn. Appl. (ICMLA),
Conf. Contemp. Comput. Informat. (ICI), Dec. Dec. 2021, pp. 1749-1754.
2019, pp. 197-203. [40]. R. Shenoy, D. Yadav, H. Lakhotiya, and J.
[33]. A. Araujo, N. Cacho, L. Bezerra, C. Vieira, and J. Sisodia, "An intelligent framework for crime
Borges, "Towards a crime hotspot detection prediction using behavioural tracking and
framework for patrol planning," in Proc. IEEE motion analysis," in Proc. Int. Conf. Emerg.
20th Int. Conf. High Perform. Comput. Smart Comput. Informat. (ESCI), Mar. 2022, pp.
Commun., IEEE 16th Int. Conf. Smart City, 1-6.
IEEE 4th Int. Conf. Data Sci. Syst. [41]. N. Aldossari, A. Algefes, F. Masmoudi, and E.
(HPCC/SmartCity/DSS), Jun. 2018, pp. 1256- Kariri, "Data science approach for crime analysis
1263. and prediction: Saudi Arabia use-case," in Proc.
[34]. A. A. Almuhanna, M. M. Alrehili, S. H. Alsubhi, 5th Int. Conf. Women Data Sci. Prince Sultan
and L. Syed, "Prediction of crime in Univ. (WiDS PSU), Mar. 2022, pp. 20-25.
neighbourhoods of New York City using spatial [42]. Y. Ma, K. Nakamura, E. Lee, and S. S.
data analysis," in Proc. 1st Int. Conf. Artif. Intell. Bhattacharyya, "EADTC: An approach to
Data Analytics (CAIDA), Apr. 2021, pp. 23-30. interpretable and accurate crime prediction," in

International Journal of Scientific Research in Science, Engineering and Technology | [Link] | Vol 11 | Issue 3 14
 P. Karthik et al Int J Sci Res Sci Eng Technol, May-June-2024, 11 (3) : 08-15

Proc. IEEE Int. Conf. Syst., Man, Cybern. (SMC),

Oct. 2022, pp. 170-177.
[43]. M. Boukabous and M. Azizi, "Multimodal
sentiment analysis using audio and text for crime
detection," in Proc. 2nd Int. Conf. Innov. Res.
Appl. Sci., Eng. Technol. (IRASET), Mar. 2022,
pp. 1-5.
[44]. L. G. A. Alves, H. V. Ribeiro, and F. A.
Rodrigues, "Crime prediction through urban
metrics and statistical learning," Phys. A, Stat.
Mech. Appl., vol. 505, pp. 435-443, Sep. 2018.
[45]. J. He and H. Zheng, "Prediction of crime rate in
urban neighborhoods based on machine
learning," Eng. Appl. Artif. Intell., vol. 106, Nov.
2021, Art. no. 104460.
[46]. H. K. R. ToppiReddy, B. Saini, and G. Mahajan,
"Crime prediction & monitoring framework
based on spatial analysis," Proc. Comput. Sci.,
vol. 132, pp. 696-705, Jan. 2018.
[47]. A. Wolf, T. R. Fanshawe, A. Sariaslan, R.
Cornish, H. Larsson, and S. Fazel, "Prediction of
violent crime on discharge from secure
psychiatric hospitals: A clinical prediction rule
(FoVOx)," Eur. Psychiatry, vol. 47, pp. 88-93,
Jan. 2018.
[48]. K. B. Sahay, B. Balachander, B. Jagadeesh, G. A.
Kumar, R. Kumar, and L. R. Parvathy, "A real
time crime scene intelligent video surveillance
systems in violence detection framework using
deep learning techniques," Comput. Electr. Eng.,
vol. 103, Oct. 2022, Art. no. 108319.
[49]. P. E. P. Utomo, "Prediction the crime
motorcycles of theft using ARIMAXTFM with
single input," in Proc. 3rd Int. Conf. Informat.
Comput. (ICIC), Oct. 2018, pp. 1-7.
[50]. V. Ingilevich and S. Ivanov, "Crime rate
prediction in the urban environment using
social factors," Proc. Comput. Sci., vol. 136, pp.
472-478, Jan. 2018.

International Journal of Scientific Research in Science, Engineering and Technology | [Link] | Vol 11 | Issue 3 15