This Article 
 Bibliographic References 
 Add to: 
Visual Human+Machine Learning
November/December 2009 (vol. 15 no. 6)
pp. 1327-1334
Raphael Fuchs, ETH Zurich
Jürgen Waser, VRVis Vienna
In this paper we describe a novel method to integrate interactive visual analysis and machine learning to support the insight generation of the user. The suggested approach combines the vast search and processing power of the computer with the superior reasoning and pattern recognition capabilities of the human user. An evolutionary search algorithm has been adapted to assist in the fuzzy logic formalization of hypotheses that aim at explaining features inside multivariate, volumetric data. Up to now, users solely rely on their knowledge and expertise when looking for explanatory theories. However, it often remains unclear whether the selected attribute ranges represent the real explanation for the feature of interest. Other selections hidden in the large number of data variables could potentially lead to similar features. Moreover, as simulation complexity grows, users are confronted with huge multidimensional data sets making it almost impossible to find meaningful hypotheses at all. We propose an interactive cycle of knowledge-based analysis and automatic hypothesis generation. Starting from initial hypotheses, created with linking and brushing, the user steers a heuristic search algorithm to look for alternative or related hypotheses. The results are analyzed in information visualization views that are linked to the volume rendering. Individual properties as well as global aggregates are visually presented to provide insight into the most relevant aspects of the generated hypotheses. This novel approach becomes computationally feasible due to a GPU implementation of the time-critical parts in the algorithm. A thorough evaluation of search times and noise sensitivity as well as a case study on data from the automotive domain substantiate the usefulness of the suggested approach.

[1] M. Ankerst, Visual Data Mining. PhD thesis, Faculty of Mathematics and Computer Science, University of Munich, 2000.
[2] D. Asimov, The grand tour: a tool for viewing multidimensional data. SIAM J. Sci Stat. Comput., 6 (1): 128–143, 1985.
[3] R. Bürger, P. Muigg, M. Ilcìk, H. Doleisch, and H. Hauser, Integrating local feature detectors in the interactive visual analysis of flow simulation data. In Proc. EuroVis 2007, pages 171–178, 2007.
[4] D. Caragea, D. Cook, H. Wickhamand, and V. Honavar, Visual methods for examining SVM classifiers. Visual Data Mining: Theory, Techniques and Tools for Visual Analytics, pages 136–153, 2008.
[5] C. Chen, Top 10 unsolved information visualization problems. IEEE Comput. Graph. Appl., 25 (4): 12–16, 2005.
[6] H. Doleisch, M. Gasser, and H. Hauser, Interactive feature specification for focus+context visualization of complex simulation data. In Proc. VisSym 2003, pages 239--248, 2003.
[7] T. M. English, Optimization is easy and learning is hard in the typical function. In Proc. Congress on Evolutionary Computation: CEC00, pages 924–931, 2000.
[8] M. Hao, U. Dayal, D. Keim, D. Morent, and J. Schneidewind, Intelligent visual analytics queries. In Proc. VAST 2007., pages 91–98, 2007.
[9] H. Hauser, R. Laramee, and H. Doleisch, Topology-based versus feature-based flow analysis - challenges and an application. In Proceedings of TopoInVis 2007: Topology-Based Methods in Visualization, 2007.
[10] A. Hertzmann, Machine learning for computer graphics: A manifesto and tutorial. In Proc. Pacific Graphics 2003, page 22. IEEE Computer Society, 2003.
[11] H. Jänicke, A. Wiebel, G. Scheuermann, and W. Kollmann, Multifield visualization using local statistical complexity. IEEE Transactions on Visualization and Computer Graphics, 13 (5): 168–175, 2007.
[12] M. Jeng, A selected history of expectation bias in physics. American Journal of Physics, 74: 578–583, 2006.
[13] J. Kehrer, F. Ladstadter, P. Muigg, H. Doleisch, A. Steiner, and H. Hauser, Hypothesis generation in climate research with interactive visual data exploration. IEEE Transactions on Visualization and Computer Graphics, 14 (6): 1579–1586, Nov.-Dec. 2008.
[14] D. Keim, F. Mansmann, J. Schneidewind, J. Thomas, and H. Ziegler, Visual analytics: Scope and challenges. Lecture Notes In Computer Science, pages 76–90, 2008.
[15] D. A. Keim, Information visualization and visual data mining. IEEE Transactions on Visualization and Computer Graphics, 8 (1): 1–8, 2002.
[16] S. Laine, Using visualization, variable selection and feature extraction to learn from industrial data. PhD thesis, TU Helsinki, 2003.
[17] R. Laramee and H. Hauser, Geometric flow visualization techniques for cfd simulation data. In Proceedings of the 21st spring conference on Computer graphics, 2005.
[18] K.-L. Ma, Machine learning to boost the next generation of visualization technology. IEEE Comput. Graph. Appl., 27 (5): 6–9, 2007.
[19] T. May and J. Kohlhammer, Towards closing the analysis gap: Visual generation of decision supporting schemes from raw data. Computer Graphics Forum, 27 (3): 911–918, 2008.
[20] W. Müller, T. Nocke, and H. Schumann, Enhancing the visualization process with principal component analysis to support the exploration of trends. In Asia Pacific Symposium on Information Visualisation, volume 60, pages 121–130, 2006.
[21] Nvidia's CUDA.
[22] F. Poulet, SVM and graphical algorithms: a cooperative approach. In ICDM '04. Fourth IEEE International Conference on Data Mining, 2004., pages 499–502, 2004.
[23] F. Poulet, Visual Data Mining, chapter Towards Effective Visual Data Mining with Cooperative Approaches, pages 389–406. Springer, 2008.
[24] P. Rautek, Semantic Visualization Mapping for Volume Illustration. PhD thesis, Vienna University of Technology, 2009.
[25] F. Rossi, Visual data mining and machine learning. In European Symposium on Artificial Neural Networks (ESANN2006), 2006.
[26] J. Seo and B. Shneiderman, A rank-by-feature framework for interactive exploration of multidimensional data. Information Visualization, 4 (2): 96–113, 2005.
[27] B. Shneiderman, Inventing discovery tools: combining information visualization with data mining. Information Visualization, 1 (1): 5–12, 2002.
[28] F.-Y. Tzeng, E. B. Lum, and K.-L. Ma, An intelligent system approach to higher-dimensional classification of volume data. IEEE Transactions on Visualization and Computer Graphics, 11 (3): 273–284, 2005.
[29] F.-Y. Tzeng and K.-L. Ma, Intelligent feature extraction and tracking for visualizing large-scale 4d flow simulations. In Proc. of the 2005 ACM/IEEE conference on Supercomputing, page 6, 2005.
[30] M. Ward, Finding needles in large-scale multivariate data haystacks. IEEE Comput. Graph. Appl., 24 (5): 16–19, 2004.
[31] D. Yang, E. A. Rundensteiner, and M. O. Ward, Analysis guided visual exploration of multivariate data. In Visual Analytics Science and Technology, VAST 2007., pages 83–90, 2007.

Index Terms:
Interactive Visual Analysis, Volumetric Data, Multiple Competing Hypotheses, Knowledge Discovery, Computer-assisted Multivariate Data Exploration, Curse of Dimensionality, Predictive Analysis, Genetic Algorithm
Raphael Fuchs, Jürgen Waser, Meister Eduard Gröller, "Visual Human+Machine Learning," IEEE Transactions on Visualization and Computer Graphics, vol. 15, no. 6, pp. 1327-1334, Nov.-Dec. 2009, doi:10.1109/TVCG.2009.199
Usage of this product signifies your acceptance of the Terms of Use.