Inspiration

  • The rapid spread of misinformation in the digital age, amplified by social media and unverified news sources, inspired us to build this tool
  • Traditional fact-checking methods are slow, often manual, and lack real-time adaptability to dynamic information flows.
  • Perplexity’s Sonar Pro API, with its real-time web search and advanced reasoning, provided a unique opportunity to build a

What it does

  • Our platform that analyzes news articles, breaks them into individual claims, and verifies them using Perplexity’s Sonar Pro API.
  • It fetches real-time articles via NewsAPI, personalizes content based on user interests, and stores data in MongoDB.
  • Then, the platform ranks sources using authenticity and tonality metrics, provides highlighted source excerpts, and generates shareable reports

How we built it

  • The platform integrates Firebase Authentication for secure email/password signup, email verification, and profile management, storing user data (country, interests) in MongoDB for personalization.
  • We use NewsAPI api-key to fetch, store and use(ONLY THE METADATA NOT ARTICLE CONTENT) in UI which are then evaluated by the users.
  • Perplexity’s Sonar Pro API powers the core processes like Claim Segmentation, Fact Analysis, Source Ranking.
  • React frontend delivers a responsive UI with dark/light mode, claim breakdowns, and KPI cards (True/False/Misleading/Unverifiable), Source evidence excerpts,
  • The feed leverages Sonar Pro’s image extraction, and multiple images are used as part of social networking feed

Challenges we ran into

  • choosing the right news provider api was initially tough & time consuming needed a lot of R&D on similar tools.
  • Implementing text highlighting for source excerpts faced cross-browser compatibility issues,

Accomplishments that we're proud of

  • Built a scalable fact-checking platform powered by Sonar Pro’s real-time web search.
  • Developed a tier-based source ranking system with authenticity and tonality metrics, enhancing transparency in source evaluation.
  • Implemented seamless text highlighting for source excerpts, improving user experience by reducing manual search time.
  • Created a collaborative feed with Sonar Pro-extracted images, driving community engagement and trust in shared reports.

What we learned

  • We learned to optimize large-context API integrations like Sonar Pro for cost and performance, balancing real-time data needs with system efficiency.
  • Navigating compliance requirements deepened our understanding of data privacy in user-driven platforms.

What's next for Fact Analyzer Platform

  • Implementing custom models & algorithms for source verification & ranking
  • expanding personalization with AI-driven topic recommendations, and supporting additional languages using Sonar Pro’s multilingual capabilities
  • We plan to introduce real-time collaboration tools for users to co-analyze articles, enhance the feed with sentiment-based filtering, and optimize image compression for faster load times.

Built With

Share this project:

Updates