Publica
Hier finden Sie wissenschaftliche Publikationen aus den FraunhoferInstituten. Interactive knowledgebased kernel PCA
 Calders, T.: Machine learning and knowledge discovery in databases. European conference, ECML PKDD 2014. Pt.2 : Nancy, France, September 1519, 2014; Proceedings Berlin: Springer, 2014 (Lecture Notes in Computer Science 8725) ISBN: 3662448505 ISBN: 9783662448502 (Print) ISBN: 9783662448519 (Online) S.501516 
 European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML/PKDD) <2014, Nancy> International Conference on Inductive Logic Programming (ILP) <24, 2014, Nancy> 

 Englisch 
 Konferenzbeitrag 
 Fraunhofer IAIS () 
Abstract
Data understanding is an iterative process in which domain experts combine their knowledge with the data at hand to explore and confirm hypotheses. One important set of tools for exploring hypotheses about data are visualizations. Often, however, traditional, unsupervised dimensionality reduction algorithms are used for visualization. These tools allow for interaction, i.e., exploring different visualizations, only by means of manipulating some technical parameters of the algorithm. Therefore, instead of being able to intuitively interact with the visualization, domain experts have to learn and argue about these technical parameters. In this paper we propose a knowledgebased kernel PCA approach that allows for intuitive interaction with data visualizations. Each embedding direction is given by a nonconvex quadratic optimization problem over an ellipsoid and has a globally optimal solution in the kernel feature space. A solution can be found in polynomial time using the algorithm presented in this paper. To facilitate direct feedback, i.e., updating the whole embedding with a sufficiently high framerate during interaction, we reduce the computational complexity further by incremental up and downdating. Our empirical evaluation demonstrates the flexibility and utility of this approach.