safeTO - A Comprehensive Community Safety Website for Torontonians

Check out our Design Document on GitHub!

Deployment

Backend

We use Gradle for dependency management. Please make sure that Gradle is installed locally on your machine.

The source code is located in the backend directory.

To build the project, run the build script using ./gradlew build for Unix-based systems or gradlew.bat build for Windows. This process will generate a .war file (currently named ROOT.war; note that this name may change in the future as we work toward a more elegant deployment solution). The .war file, which includes the compiled JAR, dependencies, and configurations, can be found in the backend/app/build/libs directory.

We recommend using Tomcat as the servlet framework. Download and extract the appropriate Tomcat release for your platform, then place the .war file in the webapps folder of the extracted Tomcat directory.

To run the application at the root level of the server, you may want to remove the default ROOT folder within the webapps directory, which contains Tomcat's welcome page. The build script is configured to generate the .war file with the name ROOT.war, ensuring it deploys at the root level.

Next, navigate to the bin directory and run startup.sh for Unix-based systems or startup.bat for Windows.

The backend application should now be accessible at http://localhost:8080. To expose the application to the public web, you can use common reverse-proxy tools like Nginx or Cloudflare Tunnel.

Since the backend API is intended to be exposed directly on the web, it's advisable to familiarize yourself with common Tomcat misconfigurations and avoid them in your production environment.

Frontend

Execute npm run build within the frontend folder to build the app for production. In CI environments such as Netlify or Cloudflare Pages, it might be advisable to set CI=false since our team is currently working on resolving several warnings. For additional information, take a look at this Stack Overflow post.

Data Access and Persistence

Our data access and persistence code is located in access directory.

We send a request to the Toronto Police Service API to fetch Major Crime Indicators Data in JSON format for the years 2019 to 2024. The fetched data is saved in the cache directory for further analysis. (Note: This directory is included in .gitignore to prevent large data files from being committed to the repository.) This approach minimizes repeated API calls and optimizes the performance of our analysis pipeline.

The analysis results are exported in JSON format to the resources directory.

Analysis

Our usecase demo is available at analysis directory.

1. Break and Enter

   • Execute the BreakAndEnterModelDemo.java file to see the analysis results for Break and Enter incidents in Toronto within a specified radius. You may adjust the latitude, longitude, radius, and threshold to customize the search parameter.
     • The results are displayed in the console, showing all known incidents within the specified radius.
     • For each incident, the following information is provided:
       • occur date: The date when the incident happened.
       • distance from you: The distance from the specified location to the incident location.
   • Additionally, we calculate the probability that Break and Enter incidents will occur more than the specified threshold in the future using a Poisson model. Based on this analysis, we provide a suggestion on whether the user is safe to live here.

2. Car Theft

   • Execute the AutoTheftSafeCaseDemo.java or AutoTheftUnsafeCaseDemo.java file to see the analysis results for auto theft incidents in Toronto within a specified radius. You may adjust the latitude, longitude, radius,threshold, and earliestYear to customize the search parameter.
   • The results are displayed in the console, showing all known incidents within the specified radius.
   • For each incident, the following information is provided:
     • occur date: The date when the incident happened.
     • distance from you: The distance from the specified location to the incident location.
   • Additionally, we calculate the probability that auto theft incidents will occur more than the specified threshold in the future using a Poisson model. we also provide a suggestion on whether is safe to park here based on our analysis.

3. Crime Data Ranking

   • Execute the NeighborhoodCrimeRankingDemo.java file to see the analysis results for crime data ranking in Toronto.
   • You may interact with the console to input the neighourhood, and sepcific crime type to get the ranking of the neighbourhood by total number of specific crime type incidents.
   • You may also leave blank for specific crime type to get the ranking of the neighbourhood by total number of all crime incidents.

Contact

Team name: Stack Underflow

• Joe Fang @MinecraftFuns [email protected]
• Minghe Ma @mhnwa [email protected]
• Bilin Nong @Bilin22 [email protected]
• Yiyun Zhang @Yiyun95788 [email protected]
• Liangyu Zhu @larryzhuly [email protected]

Some Revisions to Blueprint

1. In the User.java class, change names from List<String> to firstName, lastName;
2. In the User.java class, change userID from long to String in order to accommodate alphanumeric usernames;
3. In the User.java class, change notificationPreferences from List<String> to Map<String, Boolean>;
4. In the User.java class, change homeLocation and the elements of savedLocations from Location to SimpleLocation;
5. In the User.java class, change contacts from List<long> to List<String>;
6. In the Location.java class, apply a regular expression to Canadian postcodes;