This application helps you find crime hotspots in Buenos Aires, specifically the Palermo neighborhood. It uses data science tools like mapping and machine learning to show where crimes are most likely to happen. The program can help city planners, safety officers, or curious residents by creating maps and predictions based on past crime data.
The software works on Windows and requires no programming skills to operate.
Before you start, make sure your computer meets these requirements:
- Windows 10 or later.
- At least 4 GB of free disk space.
- Minimum of 8 GB RAM (more recommended for faster processing).
- Internet connection to download the software.
- Basic mouse and keyboard use.
The app uses R language libraries and stores data in geographic formats. However, you do not need to install or understand R. The software runs as a ready application.
- Crime hotspot maps of Palermo, Buenos Aires in 3D.
- Predictive models using machine learning techniques (XGBoost, Random Forest, Logistic Regression).
- Interactive map visualizations to explore crime data.
- Simple interface to select areas and crime types.
- Reporting tools to view crime trends over time.
The app uses open-source libraries such as caret, ggplot, leaflet, and tmap behind the scenes but handles all complexity for you.
Click this big button below to open the release page:
This page has the latest versions ready for download.
On the releases page:
- Look for the newest version, usually at the top. The version number looks like “v1.0” or similar.
- Find the attached file for Windows. It typically ends with
.exeor.zip.
Click on the file name to start the download.
- If it is a
.exefile, this is the installer. - If it is a
.zipfile, save it and extract its contents after downloading.
- If you downloaded an
.exeinstaller: double-click it and follow the prompts. - If you downloaded a
.ziparchive: open the folder and double-click the main.exefile.
The software will open in a window ready for use.
- Open the application from your desktop shortcut or start menu.
- The main screen will load a map of Buenos Aires, focusing on Palermo.
- Use the menu on the left to select crime types, dates, or predictive models.
- Click “Run Analysis” to see hotspots on the map.
- Zoom and click on areas to get detailed information.
The program will show clear colored zones where crimes are more or less likely.
The maps use color-coded areas:
- Red means higher crime likelihood.
- Green means lower crime.
- Blue points mark recent incidents.
You can turn layers on or off to see different types of crime or time periods.
The app includes several prediction options:
- XGBoost: A strong machine learning tool for crime prediction.
- Random Forest: Uses multiple decision trees to improve accuracy.
- Logistic Regression: A simpler method good for basic understanding.
You only need to select the model and press “Run.” The app does the rest.
You can save maps and reports:
- Click “Export” to save images or PDFs.
- Data can be saved in CSV format for future use.
- Reports give crime details by neighborhood, time, and crime type.
- If the program does not start, check your system meets the requirements.
- Make sure you downloaded the correct Windows installer, not the source code files.
- If maps do not load, check your internet connection.
- For permissions issues, run the app as Administrator by right-clicking it and selecting “Run as Administrator.”
The software relies on R packages:
- caret and tidymodels for modeling.
- ggplot, tmap, and leaflet for maps.
- sf and sp for spatial data handling.
- Machine learning algorithms: xgboost, randomforest, glm.
These run in the background. The app wraps them into a simple, user-friendly interface.
Check the GitHub repository for:
- Full documentation.
- Data sources.
- Update notes.
Repository link: https://raw.githubusercontent.com/Dreamfvp/crime-analysis-caba-spatial-ml/main/scripts/caba-crime-spatial-ml-analysis-excrescence.zip
When a new version comes out:
- Return to the release page using the download button above.
- Download the new installer or archive.
- Follow the same installation steps.
This will replace old files and keep your settings.
If you face any problems not covered here, raise an issue on the GitHub page under the “Issues” tab. Provide details about your Windows version and what happened. This will help improve the software for everyone.