Introduction
Analytical Platform Integrating CleverMiner is a web application wrapper around CleverMiner, a powerful engine for rules-based data mining. The platform provides a modern, user-friendly interface that makes it easy to upload datasets, run analytical tasks, and explore discovered association rules without touching the command line or Python code.
What is it?
CleverMiner is a data mining tool focused on finding patterns and rules in structured data. This platform wraps it in a full-stack web application, making it accessible to analysts and data scientists through a browser-based UI.
GUHA Procedures
The platform is built around the GUHA method (General Unary Hypotheses Automaton), a formal framework for automated hypothesis formation and data mining that generalizes standard association rules far beyond what classic apriori-based methods offer. Rather than searching for frequent itemsets, GUHA procedures work with categorical data and express richer, statistically grounded patterns of the form "If A then B with probability p".
The platform exposes all four GUHA procedures implemented in CleverMiner:
| Procedure | Purpose |
|---|---|
| 4ft-Miner | The core procedure for mining enhanced association rules. Finds rules of the form A => B with a wide range of statistical quantifiers (confidence, support, lift, and more). |
| SD4ft-Miner | Subgroup discovery variant of 4ft-Miner. Mines couples of rules to compare how a pattern behaves across two different subgroups of the data. |
| CF-Miner | Category fishery — mines rules that describe how the distribution of a target categorical attribute shifts under specific conditions. Useful for finding circumstances that boost or suppress a particular category. |
| UIC-Miner | Uplift in categories — a newer procedure designed to find conditions under which rare categories are proportionally boosted, making it especially valuable for imbalanced datasets. |
Each procedure targets a different type of analytical question, giving you a versatile toolkit for exploratory data analysis within a single platform.
The platform is built with:
- React (TypeScript) frontend
- Django backend with a REST API
- PostgreSQL for persistent storage
- Redis + Celery for background task processing
- Docker for easy deployment
Key Features
- 📂 Dataset management — upload and manage your data files
- ⚙️ Task configuration — configure and run CleverMiner analytical tasks
- 📊 Results exploration — browse and filter discovered rules and patterns
- 🐳 Docker-based deployment — get up and running with a single command
How to Access
| Service | URL |
|---|---|
| Frontend | http://localhost:3000 |
| Backend API | http://localhost:8000/api |
| API Docs | http://localhost:8000/api/docs |
Quick Start
No source code needed — just Docker.
1. Download the compose file:
curl -O https://raw.githubusercontent.com/petrstipek/Analytical-Platform-Integrating-CleverMiner/main/docker-compose.hub.yml
2. Start the application:
docker compose -f docker-compose.hub.yml up -d
3. Open http://localhost:3000, register and log in.
Make sure you have Docker Desktop installed and running before you start.
Next Steps
Once you're up and running, head over to the next sections to learn how to:
- Upload and manage your datasets
- Create and configure analytical tasks
- Explore and interpret your results