Skip to content

KikuAI-Lab/PATAS

Repository files navigation

PATAS Core

Pattern-Adaptive Transmodal Anti-Spam System

PATAS is a pattern discovery and rule management system for anti-spam operations. It analyzes historical message logs, discovers spam patterns, generates blocking rules, and evaluates their effectiveness.


How PATAS Works

PATAS uses a two-stage approach to efficiently process millions of logs:

  1. Fast Scanning (Stage 1): Processes all messages with deterministic patterns (URLs, phone numbers, keywords). This is fast and catches obvious spam cases.

  2. Deep Analysis (Stage 2): Only suspicious patterns from Stage 1 get expensive semantic analysis and AI treatment. This reduces API costs by 70-90% while maintaining high-quality pattern discovery.

The system discovers patterns by meaning (not just keywords), groups similar messages using DBSCAN clustering, and generates transparent SQL rules with precision/recall metrics.


Try It Out

Live Demo: https://patas.app/demo


Documentation

Complete documentation is available in the Wiki.

Essential for developers:


Legal Notice

Code is provided for evaluation and technical review. Any production use or long-term deployment may require a separate written agreement depending on your use case.


License

AGPL-3.0. Copyright (c) 2025 KikuAI Lab

About

Pattern-Adaptive Transmodal Anti-Spam System - On-premise deployment for messaging platforms

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages