{"id":18229,"date":"2025-05-19T17:37:35","date_gmt":"2025-05-19T15:37:35","guid":{"rendered":"https:\/\/immune.institute\/?post_type=proyecto&p=18229"},"modified":"2025-10-07T10:49:12","modified_gmt":"2025-10-07T08:49:12","slug":"prediccion-riesgos-inteligencia-artificial-distribucion-recursos","status":"publish","type":"proyectos","link":"https:\/\/immune.institute\/en\/proyectos\/prediccion-riesgos-inteligencia-artificial-distribucion-recursos\/","title":{"rendered":"Risk prediction with AI to optimise resource allocation"},"content":{"rendered":"
\n
\n

Artificial Intelligence is well known for its content generation capabilities, but it can also be used for other purposes that have a positive social impact, such as predicting high-risk crime areas.<\/p>\n\n\n\n

Thanks to the study of historical data and socio-economic, climatic or temporal variables, it is possible to predict where there is a higher risk of crime, allowing the optimisation of police resources and understanding the dynamics of crime, in this case in the city of Montevideo.<\/p>\n<\/div>\n\n\n\n

\n

Carried out by<\/mark><\/strong> Marcelo Miraballes<\/a> | Daniel Vega<\/a> | Diego Macias<\/a> | Marco Antonio Pe\u00f1a<\/a><\/p>\n\n\n\n

Qualification<\/strong> Master in<\/a> IA & <\/a>Data Science<\/a><\/p>\n\n\n\n

Technologies<\/strong> Artificial Intelligence | Machine Learning | Data Analysis and Visualisation | Optimisation of Models | Development Environments<\/p>\n<\/div>\n<\/div>\n\n\n\n

What is the motivation?<\/h2>\n\n\n\n

El auge de la recolecci\u00f3n de datos ha aumentado el inter\u00e9s en aplicar t\u00e9cnicas de Machine Learning al servicio de la seguridad p\u00fablica<\/strong>, analiz\u00e1ndolos para predecir zonas de alto riesgo delictivo y facilitando una asignaci\u00f3n eficiente de recursos.<\/p>\n\n\n\n

\n

Program aims <\/h3>\n\n\n\n
    \n
  • Predict high-risk crime areas<\/strong> using historical data and socio-economic, climatic and temporal variables.<\/li>\n\n\n\n
  • Optimising patrol resources<\/strong> through a data-driven approach.<\/li>\n\n\n\n
  • Analysing crime patterns<\/strong> to better understand the dynamics of crime in Montevideo.<\/li>\n<\/ul>\n<\/div>\n\n\n\n
    \"\"<\/figure>\n\n\n\n
    \n

    Development <\/h2>\n\n\n\n

    The project has involved the following phases:<\/p>\n\n\n\n

      \n
    • Collection of historical data and different climatic and socio-economic variables.<\/li>\n\n\n\n
    • Data cleaning and preparation, adjusting variables and dealing with missing values.<\/li>\n\n\n\n
    • Exploratory data analysis, identifying different crime patterns, analysing correlations and detecting outliers. <\/li>\n\n\n\n
    • Feature engineering, analysing the importance of different variables, encoding categorical variables and creating new variables.<\/li>\n\n\n\n
    • Predictive modelling, training the different supervised models, optimising the hyperparameters and carrying out an evaluation with performance metrics.<\/li>\n\n\n\n
    • Evaluation of the model, studying the ROC curve and overfitting, optimising the classification threshold and evaluating the metrics.<\/li>\n<\/ul>\n\n\n\n
      \"\"<\/figure>\n<\/div>\n\n\n\n

      Results <\/h2>\n\n\n\n

      The results obtained include:<\/p>\n\n\n\n

        \n
      • Robust model with an AUC above 0.75 and an F1-Score optimised by threshold adjustment techniques.<\/li>\n\n\n\n
      • Optimised model performance with the use of Optuna and efficient hyperparameter tuning.<\/li>\n<\/ul>\n\n\n\n
        \n

        El an\u00e1lisis de criminalidad en zonas cr\u00edticas para la seguridad p\u00fablica apoyado por IA y ML facilita la optimizaci\u00f3n de recursos policiales, asign\u00e1ndolos all\u00e1 donde hagan m\u00e1s falta, as\u00ed como funciona a modo de apoyo en la planificaci\u00f3n urbana y la prevenci\u00f3n de todo tipo de delitos.<\/p>\n<\/div>","protected":false},"excerpt":{"rendered":"

        Thanks to the study of historical data and socio-economic, climatic or temporal variables, it is possible to predict where there is a higher risk of crime, allowing the optimisation of police resources and understanding the dynamics of crime, in this case in the city of Montevideo.<\/p>","protected":false},"featured_media":18233,"menu_order":0,"template":"","area":[241],"class_list":["post-18229","proyectos","type-proyectos","status-publish","has-post-thumbnail","hentry","area-data"],"acf":[],"_links":{"self":[{"href":"https:\/\/immune.institute\/en\/wp-json\/wp\/v2\/proyectos\/18229","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/immune.institute\/en\/wp-json\/wp\/v2\/proyectos"}],"about":[{"href":"https:\/\/immune.institute\/en\/wp-json\/wp\/v2\/types\/proyectos"}],"version-history":[{"count":0,"href":"https:\/\/immune.institute\/en\/wp-json\/wp\/v2\/proyectos\/18229\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/immune.institute\/en\/wp-json\/wp\/v2\/media\/18233"}],"wp:attachment":[{"href":"https:\/\/immune.institute\/en\/wp-json\/wp\/v2\/media?parent=18229"}],"wp:term":[{"taxonomy":"area","embeddable":true,"href":"https:\/\/immune.institute\/en\/wp-json\/wp\/v2\/area?post=18229"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}