{"title":"An Application for Risk of Crime Prediction Using Machine Learning","authors":"Luis Fonseca, Filipe Cabral Pinto, Susana Sargento","volume":170,"journal":"International Journal of Computer and Systems Engineering","pagesStart":166,"pagesEnd":175,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/10011881","abstract":"The increase of the world population, especially in
\r\nlarge urban centers, has resulted in new challenges particularly
\r\nwith the control and optimization of public safety. Thus, in the
\r\npresent work, a solution is proposed for the prediction of criminal
\r\noccurrences in a city based on historical data of incidents and
\r\ndemographic information. The entire research and implementation
\r\nwill be presented start with the data collection from its original
\r\nsource, the treatment and transformations applied to them, choice and
\r\nthe evaluation and implementation of the Machine Learning model up
\r\nto the application layer. Classification models will be implemented to
\r\npredict criminal risk for a given time interval and location. Machine
\r\nLearning algorithms such as Random Forest, Neural Networks,
\r\nK-Nearest Neighbors and Logistic Regression will be used to predict
\r\noccurrences, and their performance will be compared according
\r\nto the data processing and transformation used. The results show
\r\nthat the use of Machine Learning techniques helps to anticipate
\r\ncriminal occurrences, which contributed to the reinforcement of
\r\npublic security. Finally, the models were implemented on a platform
\r\nthat will provide an API to enable other entities to make requests for
\r\npredictions in real-time. An application will also be presented where
\r\nit is possible to show criminal predictions visually.","references":"[1] U. Nations, \u201c68% of the world population projected to live in urban\r\nareas by 2050, says un \u2014 un desa \u2014 united nations department of\r\neconomic and social affairs,\u201d https:\/\/www.un.org\/development\/desa\/en\/\r\nnews\/population\/2018-revision-of-world-urbanization-prospects.html,\r\nMay 2018, (Accessed on 26\/10\/2019).\r\n[2] G. T. Database, \u201cIncidents over time,\u201d https:\/\/www.start.umd.edu\/gtd\/,\r\n12 2018, (Accessed on 04\/05\/2020).\r\n[3] Y. Wu, W. Zhang, J. Shen, Z. Mo, and Y. Peng, \u201cSmart city with\r\nChinese characteristics against the background of big data: Idea, action\r\nand risk,\u201d Journal of Cleaner Production, vol. 173, pp. 60\u201366, 2018.\r\n[Online]. Available: https:\/\/doi.org\/10.1016\/j.jclepro.2017.01.047\r\n[4] M. Mohammadi and A. Al-Fuqaha, \u201cEnabling Cognitive Smart Cities\r\nUsing Big Data and Machine Learning: Approaches and Challenges,\u201d\r\nIEEE Communications Magazine, vol. 56, no. 2, pp. 94\u2013101, 2018.\r\n[5] M. S. Mahdavinejad, M. Rezvan, M. Barekatain, P. Adibi, P. Barnaghi,\r\nand A. P. Sheth, \u201cMachine learning for internet of things data analysis:\r\na survey,\u201d Digital Communications and Networks, vol. 4, no. 3, pp.\r\n161\u2013175, 2018.\r\n[6] L. McClendon and N. Meghanathan, \u201cUsing Machine Learning\r\nAlgorithms to Analyze Crime Data,\u201d Machine Learning and\r\nApplications: An International Journal, vol. 2, no. 1, pp. 1\u201312, 2015.\r\n[7] Y. L. Lin, T. Y. Chen, and L. C. Yu, \u201cUsing Machine Learning to Assist\r\nCrime Prevention,\u201d Proceedings - 2017 6th IIAI International Congress\r\non Advanced Applied Informatics, IIAI-AAI 2017, pp. 1029\u20131030, 2017.\r\n[8] S. K. Rumi, K. Deng, and F. D. Salim, \u201cCrime event prediction with\r\ndynamic features,\u201d EPJ Data Science, vol. 7, no. 1, 2018. [Online].\r\nAvailable: http:\/\/dx.doi.org\/10.1140\/epjds\/s13688-018-0171-7\r\n[9] DataSF, \u201cDatasf \u2014 office of the chief data officer \u2014 city and county of\r\nsan francisco,\u201d https:\/\/datasf.org\/, 10 2020, (Accessed on 10\/12\/2020).\r\n[10] M. R. Berthold and K. P. Huber, \u201cMISSING VALUES AND\r\nLEARNING OF FUZZY RULES,\u201d vol. 6, no. 1998, pp. 171\u2013178, 1998.\r\n[11] DataSF, \u201cAnalysis neighborhoods - 2010 census tracts assigned\r\nto neighborhoods \u2014 datasf \u2014 city and county of san francisco,\u201d\r\nhttps:\/\/data.sfgov.org\/Geographic-Locations-and-Boundaries\/\r\nAnalysis-Neighborhoods-2010-census-tracts-assigned\/bwbp-wk3r\/,\r\n10 2020, (Accessed on 10\/12\/2020).\r\n[12] U. S. Census, \u201chttps:\/\/data.census.gov\/cedsci\/table?q=san francisco\r\nincome,\u201d https:\/\/data.census.gov\/cedsci\/table?q=san%20francisco%\r\n20income, 10 2020, (Accessed on 10\/12\/2020). [13] \u2014\u2014, \u201chttps:\/\/data.census.gov\/cedsci\/table?q=san francisco\r\nage&tid=acsst1y2019.s0101,\u201d https:\/\/data.census.gov\/cedsci\/table?\r\nq=san%20francisco%20age&tid=ACSST1Y2019.S0101, 10 2020,\r\n(Accessed on 10\/12\/2020).\r\n[14] \u2014\u2014, \u201chttps:\/\/data.census.gov\/cedsci\/table?q=san francisco\r\npopulation&tid=acsdp1y2019.dp05,\u201d https:\/\/data.census.gov\/cedsci\/\r\ntable?q=san%20francisco%20population&tid=ACSDP1Y2019.DP05,\r\n10 2020, (Accessed on 10\/12\/2020).\r\n[15] L. A. Shalabi, R. Mahmod, A. Azim, A. Ghani, and Y. M. Saman,\r\n\u201cA New Model for Extracting a Classifactory Knowledge from Large\r\nDatasets Using Rough Set Approach A New Model For Extracting\r\nA Classifactory Knowledge From Large Datasets Using Rough Set\r\nApproach,\u201d no. January 1999, 1999.\r\n[16] S. Learn, \u201c6.3. preprocessing data \u2014 scikit-learn 0.23.2 documentation,\u201d\r\nhttps:\/\/scikit-learn.org\/stable\/modules\/preprocessing.html, 10 2020,\r\n(Accessed on 10\/12\/2020).\r\n[17] imbalanced-learn API, \u201cimbalanced-learn api \u2014 imbalanced-learn\r\n0.5.0 documentation,\u201d https:\/\/imbalanced-learn.readthedocs.io\/en\/stable\/\r\napi.html, 10 2020, (Accessed on 10\/12\/2020).\r\n[18] S. Learn, \u201c3.1. cross-validation: evaluating estimator performance\r\n\u2014 scikit-learn 0.23.2 documentation,\u201d https:\/\/scikit-learn.org\/stable\/\r\nmodules\/cross validation.html, 10 2020, (Accessed on 10\/12\/2020).","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 170, 2021"}