lerryellis/VeriGhana-Core

An automated verification artifact designed to combat information disorder in Ghana. This system centralizes verified data from the Ministry of Information and trusted media outlets into a national fact database, using Artificial Intelligence to authenticate viral social media content and parody accounts. # VeriGhana-Core An automated verification artifact designed to combat information disorder in Ghana. This system centralizes verified data from the Ministry of Information and trusted media outlets into a national fact database, using Artificial Intelligence to authenticate viral social media content and parody accounts. VeriGhana-Core: A Centralized Automated Verification Artifact 1. Project Overview VeriGhana-Core is the technical instantiation of a research project titled "Designing a Centralized Automated Verification Artifact for Combating Information Disorder in Ghana." This project is submitted to the Department of Computer Sciences at the Ghana Institute of Management and Public Administration School of Technology in partial fulfillment of the requirements for a Bachelor of Science in Computer Science. The primary objective is to build a functional artifact that addresses the practical problem of digital misinformation, specifically targeting "Fake News Cards" and "Parody Accounts" that mislead the Ghanaian public. 2. Key Features National Fact Database: A centralized repository that aggregates official press releases and news from trusted sources (Ministry of Information, Citi Newsroom, Joy Online). Automated Ingestion Pipeline: Uses GitHub Actions to run Python-based scrapers every six hours, ensuring the database is always current. Artificial Intelligence Verification: Leverages Large Language Models (Google Gemini) and Vector Search to compare social media claims against verified ground truth. Corroboration Scoring System: Provides a visual "Truth Meter" (0-100%) to help journalists and researchers distinguish between authentic media assets and malicious forgeries. 3. Technology Stack This project utilizes a modern data stack designed for speed, scalability, and automation: Programming Language: Python 3.12 Database: Supabase (PostgreSQL with pgvector extension) Artificial Intelligence: Google Gemini 1.5 Flash (via Google AI Studio) Automat …