ExposéDuck

Tweet analysis
Tweet analysis
Insights
Insights

Inspiration

Exposé Duck was inspired by the need to give women and marginalized groups the power to avoid hostile or toxic content online, especially on platforms like Twitter. We wanted to create a tool that analyzes tweets for harmful language, such as misogyny and racism, and gamifies the experience to keep users engaged. Our aim is to help users build safer, more enjoyable online spaces while raising awareness about the prevalence of harmful content.

What it does

Exposé Duck is a Chrome extension that scans tweets, identifies and categorizes harmful content (such as misogynistic, racist, or generally toxic language), and allows users to share these findings with their network. By gamifying the process, Exposé Duck lets users level up and gain recognition for tagging and categorizing harmful content, making social media safer and more positive for everyone.

How we built it

We developed Exposé Duck using HTML, CSS, and JavaScript, with the Chrome Extension framework and manifest.json for extension configuration. For language processing and tweet analysis, we integrated the Gemini API, which allows us to detect and categorize harmful content accurately. Each component was designed to integrate seamlessly with Twitter’s UI, offering users an intuitive, accessible experience.

Challenges we ran into

One major challenge was ensuring accurate detection and categorization of toxic language, especially when dealing with sarcasm or coded language. We worked hard to balance precision and user experience, avoiding both false positives and missed content. Making the extension seamlessly integrate with Twitter’s UI also required careful work within the Chrome Extension framework.

Accomplishments that we're proud of

We’re proud of creating a functional tool that empowers users, particularly women, to engage with social media more safely. The gamification aspect is another highlight, as it encourages community engagement and adds a fun, competitive element to a serious topic. Developing a reliable, real-time categorization system for harmful content was a major technical achievement, and we’re thrilled with the initial results.

What we learned

This project taught us a great deal about natural language processing, especially in analyzing potentially hostile or harmful language. We also gained insights into user experience design within the Chrome Extension environment, ensuring Exposé Duck is both effective and enjoyable to use. Additionally, this project reinforced the importance of creating digital spaces that are inclusive and safe for all users.

What's next for Exposé Duck

Our next steps for Exposé Duck include refining our content detection algorithms, expanding the gamification features, and incorporating user feedback to improve the overall experience. We’re also planning to explore integration with additional social media platforms, making Exposé Duck a multi-platform tool for fostering safer online interactions.