ElectConnect

Onboarding Page pt.2
Onboarding Page pt.1
Tailored News Articles
Additional Features of Our App
Live Stream and Fact Check

Inspiration

Our inspiration comes from a close friend who, after recently obtaining citizenship, faced confusion about how and where to vote due to limited political involvement while on a green card. We wanted to create a solution to make voting accessible and engaging for others in similar situations.

What it does

ElectConnect is a mobile app that offers secure, personalized voting information, allowing users to register, engage in online town halls, and stay informed with tailored news updates, all designed to increase voter participation and civic engagement.

How we built it

Our ElectConnect app combines Swift and SwiftUI for the iOS frontend with Python and Flask powering the backend. We created an intuitive interface that delivers personalized voting information through a combination of HTML/CSS for our web components and UIKit for native mobile elements.

For data processing, we leveraged PyTorch and TensorFlow to analyze voting patterns and user preferences, while integrating NewsAPI, Gemini API, and Newspaper3k to aggregate real-time election information. Using NLTK, we developed an algorithm that helped us transform complex political content into accessible formats. During development, we used Matplotlib to create confusion matrices for our political bias model (using torch) evaluating our recommendation algorithm's performance, ensuring accurate information delivery to underrepresented voters.

Challenges we ran into

Throughout development, we faced several hurdles including difficulties with isolated connections (the corporate wifi) between system components, achieving consistent video streaming performance for candidate town halls, coordinating integration between our frontend and the multiple backend components, and implementing reliable web scraping functionality to gather accurate electoral articles across diverse news sources with varying formats.

Accomplishments that we're proud of

We're particularly proud of successfully integrating a comprehensive set of cohesive features that work together seamlessly, developing a smooth and remarkably fast video streaming capability for candidate interactions, implementing complex and sophisticated machine learning algorithms that personalize content for voters from diverse backgrounds, and creating an intuitive, user-friendly interface that makes the electoral process accessible even to those with limited technical experience or political knowledge.

What we learned

Throughout building ElectConnect, we learned how to design and manage custom data pipelines that aggregate and personalize election content. We developed skills in cross-platform integration, combining mobile and web technologies for a seamless user experience. Most importantly, we explored how to responsibly apply advanced AI models—balancing performance and fairness—to deliver meaningful political insights that empower underrepresented voters without bias

What's next for ElectConnect

Our vision for ElectConnect's future includes expanding our platform with comprehensive Candidate Profiles & Following capabilities where users can follow candidates they're interested in and access detailed biographies, platform summaries, and party affiliations—potentially grouped with running mates for a complete picture. We'll enhance our Content & Analysis features by aggregating and analyzing candidate posts, articles, and public statements using sophisticated AI and NLP technologies to evaluate political bias, measure centrality versus radicalism, and provide reliable fact-checking services. Additionally, we'll implement Personalized Tracking functionality allowing users to save candidates to their profiles, receive customized live news updates based on their interests, and cross-reference platforms to help them make truly informed voting decisions.