Publication detail

Intuitive and Interactive Mining

CHMELAŘ, P. STRYKA, L.

Original Title

Intuitive and Interactive Mining

Type

conference paper

Language

English

Original Abstract

In this paper, we propose a framework, internally called OLAM SE, for interactive and intuitive mining of multilevel association, characterization and classification rules. This framework is proposed as an extension of OLAP or an alternative to Han's OLAM. OLAM processes data stored in data cubes structure of which is based on a given conceptual hierarchy. OLAM SE determines minimum support value from user defined cover of data with usage of entropy coding principle. It also automatically determines the maximum threshold to avoid explaining obvious. Major part of data is thus described by frequent patterns. The presentation of results is inspired by UML diagram notation. It is a graph nodes of which are frequent data sets represented as packages including sub packages - data classes or items. Edges represent relations or patterns between packages. These patterns can be interactively explored by the user, who gets a detailed view of attractive ones. Other, possibly interesting, sets are intuitively offered to her. This is well applicable for the characterization and non-naive Bayesian classification.

Keywords

dolování dat, OLAP, OLAM, interaktivní, intuitivní, podpora, pokrytí, zřejmost.

Authors

CHMELAŘ, P.; STRYKA, L.

RIV year

2007

Released

1. 2. 2007

Publisher

VŠB-Technical University of Ostrava

Location

Ostrava

ISBN

978-80-248-1279-3

Book

ZNALOSTI 2007, Proceedings of the 6th annual conference

Pages from

308

Pages to

311

Pages count

4

BibTex

@inproceedings{BUT28581,
  author="Petr {Chmelař} and Lukáš {Stryka}",
  title="Intuitive and Interactive Mining",
  booktitle="ZNALOSTI 2007, Proceedings of the 6th annual conference",
  year="2007",
  pages="308--311",
  publisher="VŠB-Technical University of Ostrava",
  address="Ostrava",
  isbn="978-80-248-1279-3"
}