Inspiration
The inspiration for Street-Smart ultimately arose from a team discussion we had, where we really talked about the difficulty of traveling through familiar or unfamiliar areas that are associated with crime. We wanted to create something that helps improve the current situation of such travel respectively.
What it does
The web application that we designed uses a machine learning model based on a Kaggle dataset to group real world crime statistics based on location to determine what areas of the U.S. have higher or lower crime densities. Based on that information, users can reroute themselves to travel in safer areas in order to prioritize their individual safety accordingly.
How we built it
We used Python and Flask for the backend technologies. We used JavaScript, HTML/CSS, and Three.js for the frontend technologies. For proof of concept, we used a Kaggle dataset focusing on crime in the Los Angeles area. The other big piece of data that we referenced is Google Maps API, as that provided the locational data and helped us track the different paths from point A to point B when we factored in the geographical safety of a certain place.
Challenges we ran into
One of the biggest hurdles was integrating real-time crime data with geographical information from the Google Maps API. Ensuring that the routing dynamically accounted for safety while providing an efficient path was a complex task.
Accomplishments that we're proud of
We are proud of the final user interface that we came up with as well as some of the features implemented as a part of the site, such as the 3D globe, interactive elements, and supportive chatbot to name a few. Additionally, integrating the backend (Python/Flask) with the frontend (JavaScript, HTML/CSS, and Three.js) posed some synchronization issues, especially when rendering the 3D globe and interactive elements. Coordinating these components to work seamlessly together involved troubleshooting several bugs related to data flow and UI responsiveness.
What we learned
We learned that it's important to keep working as many times as possible looking at a different perspective can help us find a better way to solve it or approach it altogether. For instance, we interacted quite often with the sponsors in order to understand the technologies in more detail to figure out any issues.
What's next for Street Smart
In the future, we plan to expand Street-Smart by incorporating additional data sources beyond the Kaggle dataset, such as real-time crime reports from local law enforcement agencies and community-based safety platforms. This would allow users to receive up-to-date information and further enhance the accuracy of the safety routing. Additionally, we envision adding a crowdsourcing feature, where users can report incidents and safety concerns in real-time, which would further enrich the data and help others avoid potentially dangerous areas. Enhancing the chatbot’s capabilities to provide more proactive safety alerts and advice is also on the roadmap.
Log in or sign up for Devpost to join the conversation.