K. Morik | TU Dortmund (original) (raw)

Papers by K. Morik

Zeitschrift für wirtschaftlichen Fabrikbetrieb, 2010

Kurzfassung Das Prinzip der autonomen Automation wird als Methode der Lean Production zur Kostens... more Kurzfassung Das Prinzip der autonomen Automation wird als Methode der Lean Production zur Kostensenkung durch Vermeidung von Ausschuss und Nacharbeit genutzt. Während in Arbeitssystemen geringer Komplexität integrierte automatische Qualitätsprüfungen bereits erfolgreich zum Einsatz kommen, ist eine Übertragung auf komplexe, verkettete Arbeitssysteme derzeit nicht vollständig umsetzbar. In diesem Beitrag wird am Beispiel eines Walzprozesses ein auf maschinellen Lernverfahren basierender Ansatz zur Realisierung der autonomen Automation in komplexen, verketteten Produktionsprozessen vorgestellt. Hierbei werden verschiedene Herausforderungen, welche durch produktionstechnische Restriktionen entstehen, aufgezeigt und die Notwendigkeit von Weiterentwicklungen bestehender Lernverfahren unter der Zielsetzung der Einbindung in die Produktion dargelegt.

Lecture Notes in Computer Science, 2012

Studies in Computational Intelligence, 2011

Finding relevant subspaces in very high-dimensional data is a challenging task not only for micro... more Finding relevant subspaces in very high-dimensional data is a challenging task not only for microarray data. The selection of features is to enhance the classification performance, but on the other hand the feature selection must be stable, i.e., the set of features selected should not change when using different subsets of a population. ensemble methods have succeeded in the increase of stability and classification accuracy. However, their runtime prevents them from scaling up to real-world applications. We propose two methods which enhance correlation-based feature selection such that the stability of feature selection comes with little or even no extra runtime. We show the efficiency of the algorithms analytically and empirically on a wide range of datasets. c (5.2) with per-class-variance σ 2 c and n c the number of examples in class c ∈ {1,...,C}. The redundancy of a numerical feature set is measured by the absolute value of Pearson's correlation coefficient

Lecture Notes in Computer Science, 1999

Designing the representation languages for the input and output of a learning algorithm is the ha... more Designing the representation languages for the input and output of a learning algorithm is the hardest task within machine learning applications. Transforming the given representation of observations into a well-suited language L E may ease learning such that a simple and efficient learning algorithm can solve the learning problem. Learnability is defined with respect to the representation of the output of learning, L H . If the predictive accuracy is the only criterion for the success of learning, the choice of L H means to find the hypothesis space with most easily learnable concepts, which contains the solution. Additional criteria for the success of learning such as comprehensibility and embeddedness may ask for transformations of L H such that users can easily interpret and other systems can easily exploit the learning results. Designing a language L H that is optimal with respect to all the criteria is too difficult a task. Instead, we design families of representations, where each family member is well suited for a particular set of requirements, and implement transformations between the representations. In this paper, we discuss a representation family of Horn logic. Work on tailoring representations is illustrated by a robot application.

Applied Artificial Intelligence, 1994

Building up a knowledge base is a complex task in which theoretical knowledge needs to be integra... more Building up a knowledge base is a complex task in which theoretical knowledge needs to be integrated with practical experience. This integration can be supported by a system that can manage linking between rules, representing experts or textbook or theoretical knowledge and facts (or data), representing cases from practice. Conflicts between rules and real-world cases can have diverse causes. Case

Intelligent Technologies for Information Analysis, 2004

When learning from very large databases, the reduction of complexity is extremely important. Two ... more When learning from very large databases, the reduction of complexity is extremely important. Two extremes of making knowledge discovery in databases (KDD) feasible have been put forward. One extreme is to choose a very simple hypothesis language, thereby being capable of very fast learning on real-world databases. The opposite extreme is to select a small data set, thereby being able to learn very expressive (first-order logic) hypotheses. A multistrategy approach allows one to include most of these advantages and exclude most of the disadvantages. Simpler learning algorithms detect hierarchies which are used to structure the hypothesis space for a more complex learning algorithm. The better structured the hypothesis space is, the better learning can prune away uninteresting or losing hypotheses and the faster it becomes. We have combined inductive logic programming (ILP) directly with a relational database management system. The ILP algorithm is controlled in a model-driven way by the user and in a data-driven way by structures that are induced by three simple learning algorithms.

Exon-level expression analyses identify MYCN and NTRK1 as

University Dortmund, Dept. Computer Science, Dortmund, Germany Institute of Computer Science, FORTH, Heraklion, Greece Department of Production and Management Engineering, Technical University of Crete, Greece Director, Pediatric Surgery Clinic, University of Crete, Heraklion, Greece

this document ..................... 9 The 2.1 2.2 2.3 2.4 MiningMart System 11 M4 - The MiningMar... more this document ..................... 9 The 2.1 2.2 2.3 2.4 MiningMart System 11 M4 - The MiningMart MetaModel ................. 11 2.1.1 The Conceptual Data Model ................ 15 2.1.2 The Relational Model .................... 15 2.1.3 The Case Model ....................... 16 The Compiler ............................. 16 The Human-Computer Interface (HCI) ............... 18 The Web Platform .......................... 21 2.4.1 The Case Base ........................ 21 2.4.2 The Business Level ...................... 22 Cases and Evaluations 25 3.1 The drug store case .......................... 25 3.2 The call center case ......................... 27 3.3 The churn prediction case ...................... 28 Exploitation and Dissemination 31 4.1 The MiningMart Websites ...................... 31 4.2 The One Day Seminar ........................ 32 4.3 Exploitation Plans .......................... 32 4.3.1 Application Service Providing ................ 33 4.3.2 Research ............................ 33 4.3.3 Embedding MiningMart into other KDD tools ...... 33 4.3.4 Product Development .................... 34 5 Related Work and Conclusions 35 CONTENTS 6 Appendix 39 6.1 Experiences with Partners from Associated States ........ 39 6.2 List of documents related to MiningMart ............. 40 6.2.1 Published Papers ....................... 40 6.2.2 Deliverables .......................... 40 6.2.3 Technical Reports ...................... 42 6.2.4 Other Documents ....................... 42 Chapter I Overview 1.1 Problems in Knowledge Discovery from Data- bases The use of very large databases has enhanced in the last years from supporting transactions to additionally reporting business trends. The interest in analyzing the data has increased. One important topic is customer ...

SIGLECopy held by FIZ Karlsruhe; available from UB/TIB Hannover / FIZ - Fachinformationszzentrum ... more SIGLECopy held by FIZ Karlsruhe; available from UB/TIB Hannover / FIZ - Fachinformationszzentrum Karlsruhe / TIB - Technische InformationsbibliothekDEGerman

SIGLETIB: RO 11 (40) / FIZ - Fachinformationszzentrum Karlsruhe / TIB - Technische Informationsbi... more SIGLETIB: RO 11 (40) / FIZ - Fachinformationszzentrum Karlsruhe / TIB - Technische InformationsbibliothekDEGerman