MIME

MIME is under redesign. The new version is a web based tool called SNIPER: Snappy Interactive Pattern ExploRation



Interactive and Manual Construction of Classification Trees

Abstract:
We propose an approach, in which a user can build tree based classification models under the assistance of a computer. We extend MIME, an existing framework for pattern discovery and exploration, with several mechanisms and visualisations for aiding a user to 1) construct new trees from scratch and 2) adapt existing trees, by showing good splitting points in the data, showing troublesome parts in the tree and automating computations of trees by standard techniques. We illustrate how our system can be used by non-classification experts to build interesting decision trees. Moreover, we provide a short experimental evaluation showing how trees generated using our tool compare to existing algorithms and their pruning techniques.

PDF
Interactive and Manual Construction of Classification Trees. by Stephen Pauwel, Sandy Moens, and Bart Goethals in Proceedings of the 23rd Belgian-Dutch Conference on Machine Learning (BENELEARN 2014), 2014.



MIME: A Framework for Interactive Visual Pattern Mining

Abstract:
We present a framework for interactive visual pattern mining. Our system enables the user to browse through the data and patterns easily and intuitively, using a toolbox consisting of interestingness measures, mining algorithms and post-processing algorithms to assist in identifying interesting patterns. By mining interactively, we enable the user to combine their subjective interestingness measure and background knowledge with a wide variety of objective measures to easily and quickly mine the most important and interesting patterns. Basically, we enable the user to become an essential part of the mining algorithm. Our demo currently applies to mining interesting itemsets and association rules, and its extension to episodes and decision trees is ongoing.

Download MIME
(Please find the qtjambi library at http://qt-jambi.org/ and the weka library at http://www.cs.waikato.ac.nz/ml/weka/, download both and place the jars (qtjambi.jar, qtjambi-yoursystem.jar and weka.jar in the mime_lib directory)

PDF
MIME: A Framework for Interactive Visual Pattern Mining by Bart Goethals, Sandy Moens and Jilles Vreeken in Proceedings of the 17th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2011 ACM
PDF
MIME: A Framework for Interactive Visual Pattern Mining by Bart Goethals, Sandy Moens and Jilles Vreeken in Proceedings of the Lecture Notes in Computer Science, Vol. 6913, 2011 Springer



Example data can be found at the FIMI repository