Suche nach „[O.] [Buchtala]“ hat 13 Publikationen gefunden

F: Angewandte WirtschaftswissenschaftenF: Maschinenbau und Mechatronik

Zeitschriftenartikel

M. Bauer, O. Buchtala, T. Horeis, R. Kern, Bernhard Sick, et al.

Technical Data Mining with Evolutionary Radial Basis Function Classiers

Applied Soft Computing, vol. 9, no. 2, pp. 765-774

2009

Abstract anzeigen

This article deals with two key problems of data mining, the automation of the data mining process and the integration of human domain experts. We show how an evolutionary algorithm (EA) can be used to optimize radial basis function (RBF) neural networks used for classification tasks. First, input features will be chosen from a set of possible input features (feature selection). Second, the number of hidden neurons is adapted (model selection). It is known that interpretable (fuzzy-type) rule sets may be extracted from RBF networks. We show how appropriate training algorithms for RBF networks and penalty terms in the fitness function of the EA may improve the understandability of the extracted rules. The properties of our approach are set out by means of two industrial application examples (process identification and quality control).

F: Maschinenbau und Mechatronik

Beitrag (Sammelband oder Tagungsband)

G. Bauer, T. Taschke, O. Buchtala, Bernhard Sick

Low Level Sensor Data based Object Classication for Automotive Safety Applications

SMCia, 07, Berlin; Offenbach

2007

ISBN: 978-3800730469

F: Maschinenbau und Mechatronik

Beitrag (Sammelband oder Tagungsband)

O. Buchtala, Bernhard Sick

Goodness of Fit: Measures for a Fuzzy Classier

Proceedings of the IEEE Symposium on Foundations of Computational Intelligence (FOCI 2007); Honolulu, HI, USA; 01.-05.04.2007

2007

F: Angewandte WirtschaftswissenschaftenF: Maschinenbau und Mechatronik

Vortrag

G. Bauer, T. Taschke, O. Buchtala, Bernhard Sick

Low Level Sensor Data based Object Classication for Automotive Safety Applications

IEEE Three-Rivers Workshop on Soft Computing in Industrial Applications (SmCia/07), Passau

2007

F: Maschinenbau und Mechatronik

Beitrag (Sammelband oder Tagungsband)

O. Buchtala, Bernhard Sick

Basic Technologies for Knowledge Transfer in Intelligent Systems

Proceedings of the 2007 IEEE Symposium on Articial Life (CI-ALife 2007); Honolulu, HI, USA; 01.-05.04.2007

2007

F: Maschinenbau und Mechatronik

Beitrag (Sammelband oder Tagungsband)

O. Buchtala, Bernhard Sick

Functional Knowledge Exchange Within an Intelligent Distributed System

Architecture of computing systems - ARCS 2007, Berlin [u.a.], vol. 4415

2007

ISBN: 978-3540712671

F: Maschinenbau und Mechatronik

Beitrag (Sammelband oder Tagungsband)

O. Buchtala, Bernhard Sick

Techniques for the Fusion of Symbolic Rules in Distributed Organic Systems

IEEE Mountain Workshop on Adaptive and Learning Systems (SMCals/06); Logan, UT, USA; 24.-26.07.2006

2006

F: Maschinenbau und Mechatronik

Beitrag (Sammelband oder Tagungsband)

M. Bauer, O. Buchtala, T. Horeis, R. Kern, Bernhard Sick, et al.

Process Identification and Quality Control with Evolutionary Optimized RBF Classiers

Proceedings of the IEEE Mid-Summer Workshop on Soft Computing in Industrial Applications (SMCia/05); Espoo, Finnland; 28.-30.06.2005

2005

F: Maschinenbau und Mechatronik

Zeitschriftenartikel

O. Buchtala, M. Kilmek, Bernhard Sick

Evolutionary Optimization of Radial Basis Function Classiers for Data Mining Applications

IEEE Transactions on Systems, Man, and Cybernetics - Part B: Cybernetics, vol. 35, no. 5, pp. 928-947

2005

Abstract anzeigen

In many data mining applications that address classification problems, feature and model selection are considered as key tasks. That is, appropriate input features of the classifier must be selected from a given (and often large) set of possible features and structure parameters of the classifier must be adapted with respect to these features and a given data set. This paper describes an evolutionary algorithm (EA) that performs feature and model selection simultaneously for radial basis function (RBF) classifiers. In order to reduce the optimization effort, various techniques are integrated that accelerate and improve the EA significantly: hybrid training of RBF networks, lazy evaluation, consideration of soft constraints by means of penalty terms, and temperature-based adaptive control of the EA. The feasibility and the benefits of the approach are demonstrated by means of four data mining problems: intrusion detection in computer networks, biometric signature verification, customer acquisition with direct marketing methods, and optimization of chemical production processes. It is shown that, compared to earlier EA-based RBF optimization techniques, the runtime is reduced by up to 99% while error rates are lowered by up to 86%, depending on the application. The algorithm is independent of specific applications so that many ideas and solutions can be transferred to other classifier paradigms.

F: Maschinenbau und Mechatronik

Beitrag (Sammelband oder Tagungsband)

O. Buchtala, W. Grass, A. Hofmann, Bernhard Sick

A Distributed Intrusion Detection Architecture With Organic Behavior

Proceedings of the First CRIS International Workshop on Critical Information Infrastructures (CIIW'05); Linköping, Schweden; 17.-18.05.2005

2005

F: Elektrotechnik und MedientechnikF: Maschinenbau und Mechatronik

Vortrag

M. Bauer, O. Buchtala, T. Horeis, R. Kern, Bernhard Sick, et al.

Process Identification and Quality Control with Evolutionary Optimized RBF Classiers

IEEE Mid-Summer Workshop on Soft Computing in Industrial Applications (SMCia/05), Espoo, Finnland

2005

F: Maschinenbau und Mechatronik

Beitrag (Sammelband oder Tagungsband)

O. Buchtala, P. Neumann, Bernhard Sick

A Strategy for an Efficient Training of Radial Basis Function Networks for Classication Applications

Proceedings of the IEEE-INNS International Joint Conference on Neural Networks (IJCNN 2003); Volume 2; Portland, OR, USA; 20.-24.07.2003

2003

F: Maschinenbau und Mechatronik

Beitrag (Sammelband oder Tagungsband)

O. Buchtala, A. Hofmann, Bernhard Sick

Fast and Efficient Training of RBF Networks

Artificial neural networks and neural information processing, Berlin; Heidelberg; New York; Hong Kong; London; Milan; Paris; Tokyo, vol. Vol. 2714

2003

ISBN: 978-3540404088

Abstract anzeigen

Radial basis function (RBF) networks are used in many applications, e.g. for pattern classification or nonlinear regression. Typically, either stochastic, iterative training algorithms (e.g. gradient-based or second-order techniques) or clustering methods in combination with a linear optimisation technique (e.g. c-means and singular value decomposition for a linear least-squares problem) are applied to find the parameters (centres, radii and weights) of an RBF network. This article points out the advantages of a combination of the two approaches and describes a modification of the standard c-means algorithm that leads to a linear least-squares problem for which solvability can be guaranteed. The first idea may lead to significant improvements concerning the training time as well as the approximation and generalisation properties of the networks. In the particular application problem investigated here (intrusion detection in computer networks), the overall training time could be reduced by about 29% and the error rate could be reduced by about 74%. The second idea rises the reliability of the training procedure at no additional costs (regarding both, run time and quality of results).