Predicting the Presence of Serious Coronary Artery Disease Based on 24 Hour Holter ECG Monitoring

Fundamenta Informaticae

Buregwa-Czuma

Jankowski

2013

Self Cite

This paper investigates the approaches to an improvement of classifiers quality through the application of a domain knowledge. The expertise may be utilizable on several levels of decision algorithms such as: feature extraction, feature selection, a definition of temporal patterns used in an approximation of the concepts, especially of the complex spatio-temporal ones, an assignment of an object to the concept and a measurement of the objects similarity. The domain knowledge incorporation results then in the reduction of the size of searched spaces. The work constitutes an overview of classifier building methods efficiently utilizing the expertise, worked out latterly by Professor Andrzej Skowron research group. The methods using domain knowledge intended to enhance the quality of classic classifiers, to identify the behavioral patterns and for automatic planning are discussed. Finally it answers a question whether the methods satisfy the hopes vested in them and indicates the directions for future development.

Section: Improving the Classical Methods Of Classifier Constructionmentioning

confidence: 99%

A Domain Knowledge as a Tool For Improving Classifiers

Fundamenta Informaticae

Buregwa-Czuma

Jankowski

2013

Self Cite

“…We consider the classifier that is based on the so-called decision tree of the local discretization (see, e.g., [7], [6], [16]). It is a binary tree, created by multiple binary partitions of the set of objects into two groups (e.g., cases, states, processes, patients, observations, vehicles) with the value of a selected attribute.…”

Section: Classifier Based On a Decision Treementioning

confidence: 99%

Classifiers for Behavioral Patterns Identification Induced from Huge Temporal Data

Szczur

Dydo

et al. 2016

Self Cite

Abstract.A new method of constructing classifiers from huge volume of temporal data is proposed in the paper. The novelty of introduced method lies in a multi-stage approach to constructing hierarchical classifiers that combines process mining, feature extraction based on temporal patterns and constructing classifiers based on a decision tree. Such an approach seems to be practical when dealing with huge volume of temporal data. As a proof of concept a system has been constructed for packet-based network traffic anomaly detection, where anomalies are represented by spatio-temporal complex concepts and called by behavioral patterns. Hierarchical classifiers constructed with the new approach turned out to be better than "flat" classifiers based directly on captured network traffic data.

“…The final set of 58 temporal patterns used for the classifier construction is shown in Table III. To evaluate our classifier we use a well-known in literature train&test method and samples of data (25530 records as learning sample and 20097 records as test sample). The classifier we use is based on the so-called decision tree of the local discretization (see, e.g., [16], [17], [18]). It is a binary tree, created by multiple binary partitions (cuts) of the set of objects into two groups with the value of a selected attribute.…”

Section: Temporal Patterns Extractionmentioning

confidence: 99%

Extracting of temporal patterns from data for hierarchical classifiers construction

Szpyrka

Szczur

2015 IEEE 2nd International Conference on Cybernetics (CYBCONF)

et al. 2015

Self Cite

A method of automatic extracting of temporal patterns from learning data for constructing hierarchical behavioral patterns based classifiers is considered in the paper. The presented approach can be used to complete the knowledge provided by experts or to discover the knowledge automatically if no expert knowledge is accessible. Formal description of temporal patterns is provided and an algorithm for automatic patterns extraction and evaluation is described. A system for packet-based network traffic anomaly detection is used to illustrate the considered ideas.