Full-text resources of PSJD and other databases are now available in the new Library of Science.
Visit https://bibliotekanauki.pl
Preferences help
Polish version English version Language
enabled [disable] Abstract
Number of results

Results found: 3

Number of results on page
first rewind previous Page / 1 next fast forward last

Search results

help Sort By:

help Limit search:
first rewind previous Page / 1 next fast forward last
1
Content available remote

Entropy Based Trees to Support Decision Making for Customer Churn Management

100%
Gajowniczek K., Orłowski A., Ząbkowski T.
Acta Physica Polonica A
|
2016
|
vol. 129
|
issue 5
971-979
EN
In this work we analyze empirically customer churn problem from a physical point of view to provide objective, data driven and significant answers to support decision making process in business application. In particular, we explore different entropy measures applied to decision trees and assess their performance from the business perspective using set of model quality measures often used in business practice. Additionally, the decision trees are compared with logistic regression and two machine learning methods - neural networks and support vector machines.
2
Content available remote

Independent Component Analysis for Ensemble Predictors with Small Number of Models

100%
Szupiluk R., Ząbkowski T., Gajowniczek K.
Acta Physica Polonica A
|
2015
|
vol. 127
|
issue 3A
A-139-A-144
EN
The article presents independent component analysis (ICA) applied to the concept of ensemble predictors. The use of ICA decomposition enables to extract components with particular statistical properties that can be interpreted as destructive or constructive for the prediction. Such process can be treated as noise filtration from multivariate observation data, in which observed data consist prediction results. As a consequence of the ICA multivariate approach, the final results are combination of the primary models, what can be interpreted as aggregation step. The key issue of the presented method is the identification of noise components. For this purpose, a new method for evaluating the randomness of the signals was developed. The experimental results show that presented approach is effective for ensemble prediction taking into account different prediction criteria and even small set of models.
3
Content available remote

Q-Entropy Approach to Selecting High Income Households

88%
Gajowniczek K., Karpio K., Łukasiewicz P., Orłowski A., Ząbkowski T.
Acta Physica Polonica A
|
2015
|
vol. 127
|
issue 3A
A-38-A-44
EN
A generalized algorithm for building classification trees, based on Tsallis q-entropy, is proposed and applied to classification of Polish households with respect to their incomes. Data for 2008 are used. Quality measures for obtained trees are compared for different values of q parameter. A method of choosing the optimum tree is elaborated.
first rewind previous Page / 1 next fast forward last
JavaScript is turned off in your web browser. Turn it on to take full advantage of this site, then refresh the page.