Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Sep 7;17(9):e0274172.
doi: 10.1371/journal.pone.0274172. eCollection 2022.

Hybrid of deep learning and exponential smoothing for enhancing crime forecasting accuracy

Umair Muneer Butt  1   2 , Sukumar Letchmunan  1 , Fadratul Hafinaz Hassan  1 , Tieng Wei Koh  3
Affiliations

Hybrid of deep learning and exponential smoothing for enhancing crime forecasting accuracy

Umair Muneer Butt et al. PLoS One. .

Abstract

The continued urbanization poses several challenges for law enforcement agencies to ensure a safe and secure environment. Countries are spending a substantial amount of their budgets to control and prevent crime. However, limited efforts have been made in the crime prediction area due to the deficiency of spatiotemporal crime data. Several machine learning, deep learning, and time series analysis techniques are exploited, but accuracy issues prevail. Thus, this study proposed a Bidirectional Long Short Term Memory (Bi-LSTM) and Exponential Smoothing (ES) hybrid for crime forecasting. The proposed technique is evaluated using New York City crime data from 2010-2017. The proposed approach outperformed as compared to state-of-the-art Seasonal Autoregressive Integrated Moving Averages (SARIMA) with low Mean Absolute Percentage Error (MAPE) (0.3738, 0.3891, 0.3433,0.3964), Root Mean Square Error (RMSE)(13.146, 13.669, 13.104, 13.77), and Mean Absolute Error (MAE) (9.837, 10.896, 10.598, 10.721). Therefore, the proposed technique can help law enforcement agencies to prevent and control crime by forecasting crime patterns.

PubMed Disclaimer

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1
Fig 1. Methodology for crime forecasting.
Fig 2
Fig 2. Exponential smoothing and Bi-LSTM hybrid.
Fig 3
Fig 3. Architectural details of bi-directional LSTM and ES hybrid.
Fig 4
Fig 4. Training, validation, and testing for time-series forecasting.
Fig 5
Fig 5. Sliding window strategy for selecting the best input.
Fig 6
Fig 6. Hourly loss plot between training and evaluation without including the number of crime types in ES-BiLSTM.
Fig 7
Fig 7. Hourly loss plot between training and evaluation by including the number of crime types in ES-BiLSTM.
Fig 8
Fig 8. Histogram of seasonal, hourly crime prediction without including crime types.
The blue color refers to the actual, while the orange color refers to the forecasting data.
Fig 9
Fig 9. Histogram of seasonal, hourly predictions with including crime types.
The blue line refers to the actual, while the orange line refers to the forecast data.
Fig 10
Fig 10. Comparison of the proposed approach with state-of-the-art methods.
See this image and copyright information in PMC

References

    1. Goel RK, Saunoris JW. Does variability in crimes affect other crimes? The case of international corruption and shadow economy. Applied Economics. 2019;51(3):239–258. doi: 10.1080/00036846.2018.1494378 - DOI
    1. Estrada MAR, Ndoma I. How crime affects economic performance: The case of Guatemala. Journal of Policy Modeling. 2014;36(5):867–882. doi: 10.1016/j.jpolmod.2014.09.002 - DOI
    1. Rudolph M, Starke P. How does the welfare state reduce crime? The effect of program characteristics and decommodification across 18 OECD-countries. Journal of Criminal Justice. 2020;68:101684. doi: 10.1016/j.jcrimjus.2020.101684 - DOI
    1. Catlett C, Cesario E, Talia D, Vinci A. Spatio-temporal crime predictions in smart cities: A data-driven approach and experiments. Pervasive and Mobile Computing. 2019;53:62–74. doi: 10.1016/j.pmcj.2019.01.003 - DOI
    1. Butt UM, Letchmunan S, Hassan FH, Ali M, Baqir A, Koh TW, et al. Spatio-Temporal Crime Predictions by Leveraging Artificial Intelligence for Citizens Security in Smart Cities. IEEE Access. 2021;9:47516–47529. doi: 10.1109/ACCESS.2021.3068306 - DOI

Publication types