Skip to content

MeghaScaria/cyber-ai-system

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

24 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

AI-Powered Cyber Safety & Fraud Intelligence Agent

An intelligent, real-time cyber safety system designed to detect, analyze, and prevent digital fraud across SMS, URLs, and online communication platforms.

🚀 Overview

With the rapid rise of phishing attacks, UPI scams, and fraudulent messages, this project aims to build an AI-driven security layer that protects users from cyber threats in real time. The system leverages Natural Language Processing (NLP), machine learning, and external threat intelligence APIs to identify suspicious and malicious activity.

🎯 Key Features

  • 🔍 Smart Message Scanner
    Detects spam, phishing, and scam messages using NLP models.

  • 🌐 URL Phishing Detection
    Identifies malicious links using feature-based ML models and external APIs.

  • 🤖 AI-Based Fraud Detection
    Classifies inputs as Safe, Suspicious, or Fraud.

  • ⚠️ Real-Time Alerts
    Notifies users instantly about potential threats and suggests actions.

  • 📊 User Risk Dashboard
    Tracks fraud history, risk score, and insights.

  • 📱 Mobile App Integration
    Enables real-time scanning of SMS and user interactions.

🧠 AI/ML Components

  • NLP Models: TF-IDF, Logistic Regression, DistilBERT (planned)
  • Phishing Detection: Feature-based ML models
  • Anomaly Detection: Isolation Forest
  • Text Processing: Regex, preprocessing pipelines

🛠️ Tech Stack

  • Frontend: Flutter
  • Backend: FastAPI (Python)
  • Database: MongoDB / Firebase
  • ML/AI: Scikit-learn, Transformers (Hugging Face)
  • APIs: Google Safe Browsing, VirusTotal, Firebase FCM

📂 Project Structure

  • ml-models/ → Data processing, feature engineering, model training
  • backend/ → FastAPI APIs and model integration
  • mobile-app/ → Flutter-based mobile application
  • docs/ → Documentation

🚀 MVP Scope

  • SMS spam/fraud detection
  • URL phishing detection
  • Basic alert system
  • Backend API integration

🌍 Impact

This system aims to enhance digital safety by proactively detecting fraud and empowering users with real-time insights, contributing to a safer online ecosystem.


About

Real-time cyber fraud detection system using NLP (BERT/TF-IDF), anomaly detection, and threat intelligence APIs.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors