I:ZI Miner - pattern mining in the browser

I:ZI Miner is a web interface to the association rule mining system LISp-Miner. Its user interface resembles a web search engine. I:ZI Miner discovers multi-valued attributes, supports full range of logical connectives and 19 interest measures. An experimental relevance feedback module is used to filter rules based on previous user interactions.

Demo

The demonstration concerns fictious bank Barbora that was a part of Discovery challenge of the PKDD99 conference held in Prague in September 1999. The goal of the analysis is to find interesting segments of clients and types of loans. More information about this dataset is available here.

A quick intro to I:ZI miner can be found in the feature tutorial.

 

DISCLAIMER

I:ZI Miner is an experimental academic project. This public interface is provided only for demonstration purposes. No precautions relating to privacy and safety of connected datasets are in place. Use at your own risk.

ACKNOWLEDGMENT

The development of I:ZI Miner is suppported by the LinkedTV Project.