PL
 |
EN
Preferencje help
Polish version English version Język
Widoczny [Schowaj] Abstrakt
Liczba wyników
Powiadomienia systemowe
  • Sesja wygasła!
  • Sesja wygasła!

Znaleziono wyników: 9

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
1
Two stage ensemble selection algorithm
Baczyńska P., Burduk R.
Journal of Theoretical and Applied Computer Science
|
2015
|
Vol. 9, nr 1
3--8
EN
Multiple Classifiers Systems (MCSs) very often improve the accuracy of classification when compared with base classifiers. The building of MCSs consists of three phases: generation, selection and integration. The paper presents the two stage dynamic ensemble selection based on the analysis of the discriminant functions. The proposed in the work algorithm is applied to the binary classification tasks. In the integration phase we use the sum rule. Reported results based on the ”Pima” data set show that the proposed two stage ensemble selection is a promising method for the development of MCSs.
2
Content available Dynamical ensemble selection - experimental analysis on homogenous pool of classifiers
Baczyńska P., Burduk R.
Journal of Medical Informatics & Technologies
|
2015
|
Vol. 24
107--112
EN
The paper presents the dynamic ensemble selection based on the analysis of the decision profiles. These profiles are obtained from a posteriori probability functions returned from the base classifiers during the training process. Presented in the paper dynamic ensemble selection algorithms are dedicated to the binary classification task. In order to verify these algorithms, a number of experiments have been carried out on several medical data sets. The proposed dynamic ensemble selection is experimentally compared against the ensemble with the sum fusion method. As base classifiers we used the pool of homogeneous classifiers. The obtained results are promising because we could improve the classification accuracy of the ensemble classifier.
3
Content available The proposal of calculation classifier weights for an assembly of classifiers
Burduk R.
Journal of Medical Informatics & Technologies
|
2014
|
Vol. 23
181--186
EN
The selection of classifiers is one of the important problems in the creation of ensemble of classifiers. The paper presents the static selection in which a new method of calculating the weights of individual classifiers is used. The obtained weights can be interpreted in the context of the interval logic. It means that the particular weights will not be provided precisely but their lower and upper values will be used. A number of experiments have been carried out on several medical data sets.
4
Content available Selection of classifier in acute abdominal pain diagnosis with decision tree model
Burduk R.
Journal of Medical Informatics & Technologies
|
2011
|
Vol. 17
65--71
EN
The article presents the application of the decision tree classifier to the acute abdominal pain diagnosis. The recognition task model is based on a decision tree. In this model the decision tree structure is given by the experts. For the assumed structure of the decision tree the locally optimal strategy is considered. The problem discussed in the work shows a selection of different classifiers (parameters) to the internal nodes of the decision tree. Experiments conducted for selected medical diagnosis problem shows that the use of different parameters for k-NN classification can improve the quality of classification in comparison with the situation if it is used with the same parameter for all internal nodes of the decision tree.
5
Content available Integrated system merging the knowledge of different medical and technological disciplines WROVASC project
Burduk R., Jackowski K., Wozniak M.
Journal of Medical Informatics & Technologies
|
2010
|
Vol. 14
23--29
EN
The paper presents selected aims of WROVASK project carried on by Regional Specialist Hospital in Wroclaw with cooperation with a number of academic and research centres. The project is of interdisciplinary kind and gathers researchers and experts of number different disciplines, like chemistry, agricultural, and technical sciences, especially mechanics and informatics. Some tasks aiming at development information systems are presented and described such as: data warehouse based reporting system for the hospital, the system supporting assignment management in teleradiology with optimization of its distribution, and intelligent decision support system for automated assessment of washing surgical tools quality.
6
Content available Automatic detection and counting of platelets in microscopic image
Burduk R., Krawczyk B.
Journal of Medical Informatics & Technologies
|
2010
|
Vol. 16
173--178
EN
In this paper we present a machine learning-based approach for detecting platelet cells in microscopic smear images. Counting how many platelets appeared in each smear image is one of the basic tasks done in many laboratories. In many cases this is still done by a human — laboratory technician. Due to very small size and often great quantity of those cells, precise estimating of the number of platelets is not a trivial task. As in all man-dependent problems the whole process is very sensitive to errors, time-consuming and its accuracy is limited by human perception. We propose alternative, fully automatic solution that is free of those drawbacks. Our idea is based on the combination of techniques driven from two fields of modern computer science: the image analysis and pattern recognition ⁄ machine learning. It not only reduces the error rate, but, what is more important, also decreases the time needed for each smear image analysis. The obtained results are very satisfying and our solution is more precise than estimation based on human perception. This will improve the quality of laboratory work and allow to save time that can be spent on other important tasks.
7
Content available remote Pattern recognition of sacroileitis with the use of multistage logic with a fuzzy loss function
Burduk R.
TASK Quarterly : scientific bulletin of Academic Computer Centre in Gdansk
|
2004
|
Vol. 8, No 2
217-221
EN
The article describes the problem of pattern recognition of sacroileitis. Classification is based on a scheme of multistage recognition with a fuzzy loss function dependent on the node of the decision tree. Decision rules are based on k-nearest neighbors at particular internal nodes of the decision-tree. Paper presents influence of comparison fuzzy numbers on classifications results.
8
Content available Case of fuzzy loss function in multistage recognition algorithm
Burduk R.
Journal of Medical Informatics & Technologies
|
2003
|
Vol. 5
MI107--112
EN
The work deals with a recognition problem using a probabilistic-fuzzy model and multistage decision logic. A case where a loss function is described using fuzzy numbers has been considered. The globally optimal Bayes strategy has been calculated for this case with stage-dependent and dependent on the node of the decision tree fuzzy loss function. The obtained result is illustrated by a calculation example in which some methods for ranking fuzzy numbers were used.
9
Content available remote Concept of multistage recognition algorithm with fuzzy factors
Burduk R.
Image Processing & Communications
|
2003
|
Vol. 9, no 3-4
5-10
EN
The work deals with a recognition problem using a probabilistic-fuzzy model and multistage decision logic. A case has been considered where a value of object features is described using fuzzy numbers. The globally optimal Bayes strategy has been calculated for this case with a stage-dependent and dependent on the node of the decision tree fuzzy loss function. The obtained result is illustrated with a calculation example.
first rewind previous Strona / 1 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.