VidGram Campaign Risk Intelligence Platform

VidGram is a comprehensive end-to-end platform for influencer marketing managers to analyze influencers, understand fraud risk, estimate campaign budget loss, compare influencers, and monitor campaign risks over time.

Project Architecture

flowchart TD
    Frontend["Frontend Dashboard\n(HTML/CSS/JS)"]
    Backend1["Backend 1: Fraud Engine\n(FastAPI / Python)"]
    Backend2["Backend 2: Financial Engine\n(Express / Node)"]
    MergedData[("Merged Dataset\n(creator_dataset.json)")]
    RawData[["Raw Data\n(CSVs & JSONs)"]]

    Frontend -- "GET /api/analyze/{username}" --> Backend1
    Frontend -- "GET /api/risk/{username}?budget=..." --> Backend2
    
    Backend2 -- "Internal GET /api/analyze/{username}" --> Backend1
    
    Backend1 -- "Reads" --> MergedData
    
    RawData -- "merge_data.py" --> MergedData

Loading

The project consists of three fully integrated modules working together:

1. Frontend Dashboard A sleek, dynamic user interface built with HTML, CSS, and Vanilla JavaScript. It provides dashboards for analyzing influencers, simulating budgets, and monitoring campaigns.

2. Backend 1 - Fraud Detection Engine (Python / FastAPI) A Python service that processes raw data (profiles, posts, comments, growth metrics) from a unified creator_dataset.json file. It evaluates influencer authenticity, detects bot engagement, and identifies suspicious growth patterns.

   • Endpoint: /api/analyze/{username}

3. Backend 2 - Financial Risk Engine (Node.js / Express / TypeScript) A financial risk assessment layer that consumes data from Backend 1. It calculates business metrics such as genuine reach, fake reach, and estimated budget loss.

   • Endpoint: /api/risk/:username?budget=50000

---

🚀 Getting Started

Prerequisites

• Node.js & npm
• Python 3.9+

1. Data Processing

The platform requires a centralized JSON dataset.

# From the project root
python data_processing/merge_data.py

This generates data/merged/creator_dataset.json.

2. Start Backend 1 (Fraud Detection)

cd backend1
pip install -r requirements.txt
python app.py

Runs on http://localhost:8000

3. Start Backend 2 (Financial Risk Engine)

cd backend2/risk-engine
npm install
npx ts-node src/server.ts

Runs on http://localhost:3000

4. Start the Frontend

In a new terminal window at the project root:

python -m http.server 8080

Open your browser and navigate to: 👉 http://localhost:8080/frontend/index.html

(Note: Depending on where you start the python server, you might need to use http://localhost:8080/index.html if you started it directly inside the frontend folder.)

---

Features

• Analyze Influencer: Search for influencers (e.g., dhananjay_tech, missnidss) to see their authenticity scores and ML fraud probabilities in real-time.
• Budget Simulator: Enter an influencer and a campaign budget to project how much of your budget would be lost to fake followers.
• Campaign Monitor: View a live, automatically refreshing table of your monitored influencers, pulling real-time risk scores from the AI engines.