Wyniki wyszukiwania - BazTech

1

Can interestingness measures be usefully visualized?

Susmaga R., Szczęch I.

International Journal of Applied Mathematics and Computer Science

|

2015

|

Vol. 25, no. 2

323--336

EN

The paper presents visualization techniques for interestingness measures. The process of measure visualization provides useful insights into different domain areas of the visualized measures and thus effectively assists their comprehension and selection for different knowledge discovery tasks. Assuming a common domain form of the visualized measures, a set of contingency tables, which consists of all possible tables having the same total number of observations, is constructed. These originally four-dimensional data may be effectively represented in three dimensions using a tetrahedron-based barycentric coordinate system. At the same time, an additional, scalar function of the data (referred to as the operational function, e.g., any interestingness measure) may be rendered using colour. Throughout the paper a particular group of interestingness measures, known as confirmation measures, is used to demonstrate the capabilities of the visualization techniques. They cover a wide spectrum of possibilities, ranging from the determination of specific values (extremes, zeros, etc.) of a single measure, to the localization of pre-defined regions of interest, e.g., such domain areas for which two or more measures do not differ at all or differ the most.

2

Visualization support for the analysis of properties of interestingness measures

Susmaga R., Szczęch I.

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2015

|

Vol. 63, nr 1

315--327

EN

The paper considers a particular group of rule interestingness measures, called Bayesian confirmation measures, which have become the subject of numerous, but often exclusively theoretical studies. To assist and enhance their analysis in real-life situations, where time constraints may impede conducting such time consuming procedures, a visual technique has been introduced and described in this paper. It starts with an exhaustive and non-redundant set of contingency tables, which consists of all possible tables having the same number of observations. These data, originally 4-dimensional, may, owing to an inherent constraint, be effectively represented as a 3-dimensional tetrahedron, while an additional, scalar function of the data (e.g. a confirmation measure) may be rendered using colour. Dedicated analyses of particular colour patterns on this tetrahedron allow to promptly perceive particular properties of the visualized measures. To illustrate the introduced technique, a set of 12 popular confirmation measures has been selected and visualized. Additionally, a set of 9 popular properties has been chosen and the visual interpretations of the measures in terms of the properties have been presented.

3

Pruning discovered sequential patterns using minimum improvement threshold

Prinke S., Wojciechowski M., Zakrzewicz M.

Foundations of Computing and Decision Sciences

|

2006

|

Vol. 31, No. 1

43-57

EN

Discovery of sequential patterns is an important data mining problem with numerous applications. Sequential patterns are subsequences frequently occurring in a database of sequences of sets of items. In a basic scenario, the goal of sequential pattern mining is discovery of all patterns whose frequency exceeds a user-specified frequency threshold. The problem with such an approach is a huge number of sequential patterns which are likely to be returned for reasonable frequency thresholds. One possible solution to this problem is excluding the patterns which do not provide significantly more information than some other patterns in the result set. Two approaches falling into that category have been studied in the context of sequential patterns: discovery of maximal patterns and closed patterns. Unfortunately, the set of maximal patterns may not contain many important patterns with high frequency, and discovery of closed patterns may not reduce the number of resulting patterns for sparse datasets. Therefore, in this paper we propose and experimentally evaluate the minimum improvement criterion to be used in the post-processing phase to reduce the number of sequential patterns returned to the user. Our method is an adaptation of one of the methods previously proposed for association rules.