Skip to content

SquizAI/fiu_hack

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

localpulse

coralcamera projecttt

Inspiration

The inspiration for LocalPulse came from the desire to foster stronger, more informed, and engaged communities. We recognized a gap in the accessibility of local information - often scattered across various sources or simply not available to residents. By bringing together data on crime, small businesses, real estate trends, and more, we aimed to create a unified platform that allows individuals to feel more connected, safe, and empowered in their neighborhoods. Our goal is to transform communities through transparency, insight, and engagement.

What it does

LocalPulse is an innovative app that allows users to connect with their community by providing instant access to comprehensive data and insights. It offers crime analysis, small business details, and city historical trends, giving users a deeper understanding of their neighborhood. With features like real-time police reports, chatbot FAQs with live city information, live traffic and weather trends, real estate analytics and predictions, social media sentiment analysis, and even live camera feeds with object detection, LocalPulse provides a one-stop solution for staying informed. It also includes a community forum for local discussions, the ability to file reports, custom alerts, and much more—all designed to help residents stay aware, safe, and connected.

How we built it

We built LocalPulse using a diverse range of modern technologies that enabled us to integrate multiple data sources, implement advanced features, and create an engaging user experience. For our backend infrastructure, we used Google Cloud for hosting and data processing, leveraging Google SQL for managing our relational database. Docker was used to containerize our applications, ensuring consistent environments across all stages of development.

The data analysis and machine learning aspects were powered by pandas, NumPy, and PyTorch, allowing us to handle real-time data analysis for crime, real estate, and traffic predictions. We also used scikit-learn for building models that analyze historical data and predict trends, while PyDantic helped us maintain robust data validation.

The front end was built using Streamlit to provide an interactive user interface, supported by Plotly and Matplotlib for visualizing trends and insights in an accessible manner. Additionally, natural-language-processing features were integrated using TextBlob and Google's Generative AI to provide users with chatbot responses and social media sentiment analysis. Deep Translator was used to add multilingual support to reach a wider audience.

For live camera feeds with object detection, we utilized computer vision techniques, integrating Google Maps for geolocation tracking and GIS for spatial analysis. Langchain and Claude powered the chatbot, providing an intelligent conversational interface for users. We employed Smarthub for community forum management and API integrations to connect with various local data sources effectively.

The system design was visualized and planned using Lucidchart to ensure smooth collaboration among team members, and we used PostgreSQL as our primary database for managing community posts and historical data.

Challenges we ran into

One of the biggest challenges we faced was consolidating data from various disparate sources while ensuring its accuracy and timeliness. Integrating real-time data for crime reports, traffic trends, and social media posed technical challenges in maintaining synchronization and preventing data overload. Another challenge was ensuring the security and privacy of user information, especially when dealing with sensitive data like crime reports and user-generated posts. We also encountered some difficulties in optimizing our object detection feature to work efficiently on live video streams without significant latency.

Accomplishments that we're proud of

We're incredibly proud of the all-in-one nature of LocalPulse and how we managed to bring together so many different aspects of community life in a single platform. Successfully integrating real-time crime analysis, traffic data, and live camera feeds was a major technical achievement. We're also proud of creating a user-friendly and visually appealing interface that makes complex data easily accessible and understandable for everyone. Lastly, developing a chatbot that provides live, up-to-date information about the city was a challenging but rewarding feature that we feel will make a real difference for users.

What we learned

Throughout the development of LocalPulse, we learned a great deal about working with real-time data and integrating multiple APIs into a cohesive system. Collaboration was key, and we learned a lot about managing team responsibilities and working under tight hackathon time constraints, which pushed us to iterate quickly and make effective decisions.

What's next for LocalPulse

Moving forward, we want to expand LocalPulse by partnering with local municipalities, police departments, and community organizations to further enhance the accuracy and depth of our data. We plan to improve the machine learning models for even better predictive capabilities, particularly for real estate trends and crime pattern analysis. We also want to add more community-centric features, such as localized event announcements, volunteer opportunities, and emergency response integration. Finally, we aim to launch LocalPulse in more cities, tailoring it to meet the specific needs and dynamics of each community, making it a truly indispensable resource for neighborhoods everywhere.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages