RESEARCH FIELDS

Computational Intelligence

Machine Learning

Pattern Recognition

Intelligent Systems and Control

Data Science

Human-Machine Hybrid Intelligence

Neurocomputing and Neuroengineering

RESEARCH SUMMARY

Keywords: Computational intelligence; Machine learning; Intelligent systems and control; Intelligent data modeling and analysis; Human-machine systems; Brain-machine interaction; Brain signal processing; Modeling and control of biological systems.

My main research interests include:

  • Intelligent systems and control (Development of novel computational intelligence and machine learning techniques, such as fuzzy systems, artificial neural networks and evolutionary multi-objective optimization): In this area I focus on computational intelligence based modeling and control techniques (e.g., adaptive fuzzy systems, multi-objective evolutionary algorithms, granular computing, recurrent neural networks, etc.) and their applications in various real-world engineering, biomedical and business problems (e.g., automatic train control, structural vibration control, optimum design and processing of metals/alloys, biomedical signal processing, computational neuro-ergonomics, retail supply chain forecasting and optimization, etc.).
  • Data science and engineering (complex data mining and knowledge discovery: regression, classification and clustering): This area is concerned with construction of data-driven model from large-scale data. I have proposed advanced machine learning algorithms (deep learning, unsupervised learning, semi-supervised learning, ensemble learning, transfer learning, extreme learning machine, dynamical/adaptive machine learning, etc.) and apply them to identify and predict complex nonlinear phenomenon or mechanisms underlying big data.
  • Modeling, analysis, and control of complex biomedical systems (brain signal processing, neuroergonomics, human-machine hybrid intelligence, brain-machine interaction): In this area I have concentrated on advanced modeling, analysis, optimization, and control theory and algorithms of complex biomedical systems.


PRINCIPAL RESEARCH CONTRIBUTIONS

My core research expertise is in computational intelligence, machine learning, human-machine hybrid intelligence, neurocomputing and neuroengineering. I have produced 68 referred journal publications, 14 books and book chapters, and 68 referred conference papers. For a list of publications, please visit my Google Scholar profile.

My significant research contributions include:

1)    Intelligent techniques for industrial control systems

  • Developed a long-range predictive control method for automatic train stopping (ATS) using an associative memory type neural network based on discrete Tayler series.
  • Proposed a new fuzzy predictive control method for automatic train operation (ATO).
  • Developed a nonlinear decentralized robust control scheme for active vibration control of high-rise structures. The results showed that the proposed control scheme not only can significantly reduce the structural response under the excitation of seismic wave or strong wind, but also is robust w.r.t. hardware failures.

2)    Combination of conventional and intelligent techniques for big data analysis

  • Developed an internal model based adaptive Kalman filtering (IMAKF) technique to filter the GPS-measured vehicular position and velocity data.
  • Proposed a wavelet-based optimal denoising algorithm with high time-frequency resolution without using statistical assumptions about the signal and noise.
  • Developed new time series forecasting methods, based on an adaptive regularization network and a nonlinear adaptive fuzzy approximator, for single-trial estimation of cerebral evoked potentials.
  • Developed and applied adaptive machine learning and evolutionary multi-objective optimization algorithms to retail supply chain system modeling and optimization.

3)    Adaptive multi-modal human-machine systems and brain-machine interfaces

  • Developed an adaptive fuzzy system technique for accurate modeling and classification of Operator Functional State (OFS), Mental WorkLoad (MWL), and Cognitive Task-Load (CTL) of human operators based on the measured physiological data (e.g., neurophysiological and cardiovascular data). The work paves the way for preventing operator performance degradation or breakdown in safety-critical human-machine systems, which are ubiquitous in such domains as manned spacecraft, UAVs, air traffic control, nuclear power plant, and intelligent transportation systems.
  • Designed an adaptive human-machine cooperative control system.

PROFESSIONAL ESTEEM/RECOGNITION (selected)

Associate Editor of Frontiers in Neuroscience, 2016-

Member of Editorial Board of Cognitive Neurodynamics (Springer), Cognition, Technology and Work (Springer), and Heliyon (Elsevier), 2017-

Chair of the IFAC TC 4.5 on Human Machine Systems, 2017-

Member of the CAA (Chinese Association of Automation, the Chinese NMO of IFAC) TC on Human-Machine Hybrid Intelligence, 2017-

External Examiner/Reviewer for many universities in Asia and Europe.

Invited Keynote Speaker at the 2016 Intl. Conf. Electrical Engineering and Automation (EEA2016), Hong Kong, June 24-26, 2016.

Invited Keynote Speaker at the 2nd Intl. Conf. on Fuzzy Systems and Data Mining (FSDM2016), Macau, Dec. 11-14, 2016.

Invited Keynote Speaker at the 1st Spring Expert Meeting of Initiative Biokybernetik, Lausanne (Switzerland), Mar. 21-23, 2018.

IPC Member for numerous international conferences: 19th and 20th IFAC World Congress, ICCN2015, ICCN2017, ICANN2017, IFAC-BMS2015, CDIT2016, IFAC-HMS2016, IFAC-CPHS2016, CCC (2014-), VTITS2016, IFAC-CTS2018, IFAC-BMS2018, IFAC-CPHS2018, IFAC-ICONS2019, ......

PUBLICATIONS

140+ journal and conference papers, 4 books, and 10 book chapters (see my Google Scholar profile for the list)

List of Selected Publications

[1]    J. Zhang*, J. Xia and R. Wang, Modeling and Supervisory Control of Hybrid Dynamical Systems via Fuzzy l-complete Approximation Approach, Nonlinear Analysis: Hybrid Systems, vol. 27, pp. 390-415, 2018.

[2]    Z Yin and J Zhang, Task-generic mental fatigue recognition based on neurophysiological signals and dynamical deep extreme learning machine, Neurocomputing, vol. 283, pp. 266-281, 2018.

[3]    J. Zhang*, S. Li and R. Wang, Pattern Recognition of Momentary Mental Workload

Based on Multi-Channel Electrophysiological Data and Ensemble Convolutional Neural Networks, Front. Neurosci., vol. 11, Article 310, pp. 1-16, May 30, 2017, doi: 10.3389/fnins.2017.00310.

[4]    J. Zhang* and T. Edwards, Guest editorial for special issue on modeling and analysis of human-machine systems in transportation, Cognition, Technology & Work, vol. 19 (4), pp. 543-544, 2017.

[5]    J. Zhang*, Y. Wang, and S. Li, Cross-subject Mental Workload Classification Using Kernel Spectral Regression and Transfer Learning Techniques, Cognition, Technology & Work, vol. 19 (4), pp. 587-605, 2017.

[6]    J. Zhang* and S. Li, A Deep Learning Scheme for Mental Workload Classification based on Restricted Boltzmann Machine, Cognition, Technology & Work, vol. 19 (4), pp. 607-631, 2017.

[7]    J. Zhang*, X. Cui, J Li and R. Wang, Imbalanced Classification of Mental Workload Using a Cost-Sensitive Majority Weighted Minority Oversampling Strategy, Cognition, Technology & Work, vol. 19 (4), pp. 633-653, 2017.

[8]    Z. Yin, L. Liu, L. Liu, J. Zhang and Y. Wang, Dynamical recursive feature elimination technique for neurophysiological signal based emotion recognition, Cognition, Technology & Work, vol. 19 (4), pp. 667-685, 2017.

[9]    Z. Yin and J. Zhang, Cross-subject recognition of operator functional states via EEG and switching deep belief networks with adaptive weights, Neurocomputing, vol. 260, pp. 349-366, 2017.

[10]  J. Zhang*, Z. Yin and R. Wang, Design of An Adaptive Human Machine System Based on

Dynamical Pattern Recognition of Cognitive Task-load, Front. Neurosci., 11:129, pp. 1-18, Mar. 2017, doi: 10.3389/fnins.2017.00129

[11]  J. Zhang*, Z. Yin and R. Wang, Nonlinear Dynamic Classification of Momentary Mental Workload Using Physiological Features and NARX Model based Least-Square Support Vector Machines. IEEE Trans. on Human-Machine Systems, vol. 47(4), pp. 536-549, Aug. 2017.

[12]  J. Zhang*, J. Xia, J. M. Garibaldi, P. P. Groumpos, and R. Wang, Modeling and Control of Operator Functional State in a Unified Framework of Fuzzy Inference Petri Nets, Computer Methods and Programs in Biomedicine, vol. 144, pp. 147-163, 2017.

[13]  Z. Yin, Y. Wang, L. Liu, W. Zhang, and J. Zhang*, Cross subject EEG indicator selection for emotion recognition using transfer recursive feature 

elimination, Front. Neurorobot., 11:19, pp. 1-16, Apr. 2017, doi: 10.3389/fnbot.2017.00019

[14]  J. Zhang*, X. Liu, Z. Hu et al., Intelligent Identification of Multi-Level Nanopore Signature for Accurate Detection of Cancer Biomarkers, Chemical Communications, 2017, vol. 53, pp. 1076-1079, 2017. (IF: 6.319)

[15]  J. Zhang*, X. Liu, Y. Ying et al., High-bandwidth Nanopore Data Analysis by Using a Modified Hidden Markov Model, Nanoscale, vol. 9 (10), pp. 3458-3465, Mar. 2017 (IF: 7.76).

[16]  J. Zhang* and Zhong Yin, Pattern Classification of Instantaneous Cognitive Task-load Through

GMM Clustering, Laplacian Eigenmap and Ensemble SVMs, IEEE/ACM Trans. On

Computational Biology and Bioinformatics, vol. 14 (4), pp. 947 - 965, July-Aug. 2017. (IF: 1.955)

[17]  Z. Yin, M. Zhao, Y. Wang, J. Yang and J. Zhang, Recognition of emotions using multimodal physiological signals and an ensemble deep learning model, Computer Methods and Programs in Biomedicine, vol. 140, pp. 93-110, 2017.

[18]  Z. Yin and J. Zhang, Cross-session classification of mental workload levels using EEG and an adaptive deep learning model, Biomedical Signal Processing and Control, vol. 33, pp. 30-47, 2017.

[19]  J. Zhang* and S. Yang, A novel PSO algorithm based on an incremental-PID-controlled search

strategy, Soft Computing, vol. 20, pp. 991-1005, 2016.

[20]  J. Zhang*, S. Yang and R. Wang, Operator functional state estimation based on EEG-data-driven fuzzy model, Cognitive Neurodynamics, 10(5), 375-383, Oct. 2016.

[21]  J. Zhang*, Z. Yin and R. Wang, Recognition of Mental Workload Levels Under Complex 

Human-Machine Collaboration by Using Physiological Features and Adaptive Support Vector

 Machines, IEEE Trans. on Human-Machine Systems, vol. 45(2), pp. 200-214, Apr. 2015.

[22]  J. Zhang* and S. Yang, An incremental-PID-controlled Particle Swarm Optimization algorithm

For EEG-data-based estimation of operator functional state, Biomedical Signal Processing 

and Control, vol. 14, pp. 272-284, 2014.

[23]  Z. Yin and J. Zhang*, Identification of Temporal Variations in Mental Workload Using Locally-

Linear-Embedding-based EEG Feature Reduction and Support-Vector-Machine-based Clustering and Classification Tech-niques, Computer Methods and Programs in Biomedicine, vol. 115, pp. 119-134, 2014.

[24]  Zhong Yin and J. Zhang*, Operator Functional State Classification Using Least-Square Support Vector Machine based Recursive Feature Elimination Technique, Computer Methods and Programs in Biomedicine, vol. 113 (1), pp. 101-115, Jan. 2014.

[25]  L. Ren, B. Guo and J. Zhang et al., Mid-term efficacy of percutaneous laser disc decompression for treatment of cervical vertigo, European J. of Orthopaedic Surgery & Tramatology, vol. 24 (S1), pp. S153-S158, 2014.

[26]  L. Ren, Z. Han, J. Zhang, T. Zhang, J. Yin, X. Liang, H. Guo, and Y. Zeng, Efficacy of percutaneous laser disc decompression on lumbar spinal stenosis, Lasers in Medical Science, vol. 29(3), pp. 921-923, May 2014.

[27]  J. Zhang*, Hua Liu, Xiaodi Peng, Joerg Raisch, and Rubin Wang, Classifying human operator functional state based on electrophysiological and performance measures and fuzzy clustering

method, Cognitive Neurodynamics, vol. 7, pp. 477-494, 2013.

[28]  J. Zhang*, Panpan Qin, Joerg Raisch, and Rubin Wang, Predictive modeling of human operator cognitive state via sparse and robust support vector machines, Cognitive Neurodynamics, vol. 7 (5), pp. 395-407, 2013.

[29]  Shaozeng Yang and Jianhua Zhang*, An Adaptive Human-Machine Control System based on Multiple Fuzzy Predictive Models of Operator Functional State, Biomedical Signal Processing and Control, vol. 8 (3), pp. 302-310, May 2013.

[30]  Z.G. Gui, P.C. Zhang, J. Zhang and Y.J. Zeng, A X-ray sharpening algorithm using nonlinear module, Imaging Science Journal, vol. 60 (1), pp. 3-8, 2012.

[31]  Raofen Wang, J. Zhang, Yu Zhang, and Xingyu Wang, Assessment of human operator functional state using a novel differential evolution optimization based adaptive fuzzy model, Biomedical Signal Processing and Control, vol. 7 (5), pp. 490-498, Sept. 2012.

[32]  J. Jin, B.Z. Allison, C. Brunner, B. Wang, X. Wang and J. Zhang, P300 Chinese input system based on Bayesian LDA, Biomedizinische Technik / Biomedical Engineering, vol. 55 (1), pp. 5-18, Feb. 2010.

[33]  L. Hu, X. Xu, Y. Gong, X. Fan, L. Wang, J. Zhang and Y. Zeng, Percutaneous biphasic electrical stimulation for treatment of obstructive sleep apnea syndrome, IEEE Trans. on Biomedical Engineering, vol. 55 (1), pp. 181-187, Jan. 2008.

[34]  X. Ren, Z. Peng, Q. Zeng, C. Peng, J. Zhang et al., An improved method for Daugman’s iris localication algorithm, Computers in Biology and Medicine, vol. 38, pp. 111-115, 2008.

[35]  J. Zhang* and Johann F. Böhme, “Adaptive regularization network based neural modeling paradigm for nonlinear adaptive estimation of cerebral evoked potentials,” Medical Engineering & Physics, vol. 29, pp. 1008-1018, 2007.

[36]  Y Zeng, J Zhang, H Yin and Y Pan, Visual evoked potential estimation by adaptive noise cancellation with neural-network-based fuzzy inference system, J. of Medical Engineering & Technology, vol. 31 (3), pp. 185-190, May/June 2007.

[37]  Y.-J. Zeng, J. Zhang, B. Shen, et al., Measuring blood flow velocities based on three image processing techniques, Medical Physics, vol. 32 (4), pp. 1187-1192, 2005.

[38]  Y. Wang, S. Zhang, Y.M. Yang, Z.C. Luo, J. Zhang, and Y.J. Zeng, An analysis of frequency response for the blood flow of volume pulse in microcirculation, Bio-Medical Materials and Engineering, vol. 15 (3), pp. 189-197, 2005.

[39]  J. Zhang*, J.F. Böhme and Y.-J. Zeng, “A nonlinear adaptive fuzzy approximator technique with its application to prediction of non-stationary EEG dynamics and estimation of single-sweep evoked potentials,” Technology & Health Care, vol. 13 (1), pp. 1-21, 2005.

[40]  J. Zhang*, K. Janschek, J.F. Böhme, and Y.-J. Zeng, Multi-resolution dyadic wavelet de-noising approach for extraction of visual evoked potentials in the brain, IEE Proc.-Vision, Image & Signal Processing, vol. 151 (3), pp. 180-186, June 2004.

[41]  H.-E. Yin, Y.-J. Zeng, J. Zhang, and Y.-F. Pan, Application of adaptive noise cancellation with neural-network-based fuzzy inference system for visual evoked potentials estimation, Medical Engineering & Physics, vol. 26 (1), pp. 87-92, Jan. 2004.

[42]  J. Song, S. Zhang, Y. Diao, Z. Luo, J. Zhang et al., Predicting pregnancy-induced hypertension with dynamic hemodynamics, European J. of Obstetrics & Gynecology and Reproductive Biology, vol. 117 (2), pp. 162-168, 2004.

[43]  Q. Wang, Y.-J. Zeng, P. Luo, J.-L. Hu and J. Zhang, A specialized plug-in software module for computer-aided quantitative measurement of medical images, Medical Engineering & Physics, vol. 25 (10), pp. 887-892, Dec. 2003.

[44]  J. Zhang, Cerebral Evoked Potential Estimation: Time-frequency Analysis and Intelligent Data Modeling Methods, Berlin, Germany: dissertation.de - Verlag im Internet GmbH, 2005, 155 pp, ISBN 3-89825-974-9.

[45]  J. Zhang, L. Zhuo and Y.-H. Zhang, Digital Signal Processing, Beijing, China: Science Press, January 11, 2002, 339 pp, ISBN 7-03-009550-2. (Chinese Translated Edition)

[46]  J. Zhang, Mechatronic Systems Design: Fundamentals of Modeling, Simulation and Implementation, Beijing, China: Tsinghua University Press, Jan. 2017, 270 pp, ISBN 978-7-302-45577-6 (Chinese Translated Edition)

[47]  J. Zhang, Mechatronic Systems Design: Implementation, Control and Analysis, Beijing, China: Tsinghua University Press, May 2017, 249 pp, ISBN 978-7-302-45735-0 (Chinese Edition)

[48]  J. Zhang and R. Wang, Adaptive Fuzzy Modeling based Assessment of Operator Functional State in Complex Human-Machine Systems, in G. M. Dimirovski (Ed), Complex Systems: Relationship between Control, Communications and Computing, Chapter 9, 189-210, Springer, 2016.

[49]  J. Zhang, S. Yang, Z. Yin, and R. Wang, Estimation of Operator Functional State Using 

EEG-Data-driven Fuzzy Models with Entropy-based Partition, in R. Wang and X. Pan (Eds), Advances in Cognitive Neurodynamics (V), Chapter 69, pp. 511-518, Springer Verlag, 2016.

[50]  Z. Yin, J. Zhang, and R. Wang, Neurophysiological Features Based Detection of Mental 

Workload by Ensemble Support Vector Machines, in R. Wang and X. Pan (Eds), Advances in Cognitive Neurodynamics (V), Chapter 64, pp. 469-475, Springer, 2016.

[51]  J. Xia, J. Zhang, and R. Wang, Modeling of Adaptive Human-Machine Systems Based on Fuzzy

Inference Petri Nets, in R. Wang and X. Pan (Eds), Advances in Cognitive Neurodynamics (V), Chapter 67, pp. 493-499, Springer, 2016.

[52]  P. Chen and J. Zhang, Performance comparison of machine learning algorithms for EEG-signal-based emotion recognition, in A. Lintas, S. Rovetta, P. Verschure and A. Villa (Eds.), ICANN2017, Part I, LNCS 10613, pp. 208-216, Springer Int. Publ. AG, 2017.

[53]  J. Li and J. Zhang, Mental workload classification based on semi-supervised extreme learning machine, in A. Lintas, S. Rovetta, P. Verschure and A. Villa (Eds.), ICANN 2017, Part II, LNCS 10614, pp. 297-304, Springer Int. Publ. AG, 2017.


INVITED TALKS

l  Mental Workload Classification Using Semi-Supervised Extreme Learning Machine, Keynote lecture,  Spring Expert Meeting of Initiative Biokybernetik, Neuroheuristic Research Group (Prof Alessadro E.P. Villa, Member of the Swiss Academy of Technical Sciences (SATW)), Department of Information Systems, Université de Lausanne (UNIL), Switzerland, Mar. 21-23, 2018.

l  Prediction and Regulation of Operator Functional State Using Electrophysiological Data and Fuzzy Rule-based Models, Keynote lecture, Forum on Cognitive Neuroscience Frontiers, Apr. 26, 2017, Hangzhou, China

l  Pattern Recognition of Instantaneous Mental Workload: A Comprehensive Empirical Comparison of Different Feature Reduction and Classification Methods, Keynote lecture, Int. Conf. Electrical Engineering and Automation (EEA2016), June 24-26, 2016, Hong Kong

l  Pattern Recognition of Momentary Mental Workload: An Empirical Comparison of Different Feature Reduction and Classification Methods, Keynote lecture, Int. Conf. Artificial Intelligence Science and Technology (AIST2016), July 15-17, 2016, Shanghai, China

l  Imbalanced Data Classification Based on a Cost-Sensitive Majority Weighted Minority Oversampling Approach, Keynote Lecture, 2nd Annual Int. Conf. on Electronics, Electrical Engineering and Information Science (EEEIS2016), December 2-4, 2016, Xi'an, China

l  Mental Workload Classification Using a DySMOTE-algorithm-based Neural Network Approach, Keynote lecture, 2nd Int. Conf. on Electrical Engineering and Industrial Engineering (ICEEIE2016), Dec. 11-12, 2016, Shanghai, China

l  Prediction and Regulation of Operator Functional State Using Electrophysiological Data and Fuzzy Rule-based Models, Keynote lecture, 2nd Int. Conf. on Fuzzy Systems and Data Mining (FSDM2016), Dec. 11-14, 2016, Macau

l  A Self-Organizing Fuzzy Neural Network for Nonlinear System Control, Guest seminar, Oct. 29, 2015, Control Systems Group, Technical University of Berlin, Germany

l  Multi-class Classification of Imbalanced Dataset via A DySMOTE-based Neural Network Approach, Guest seminar, Guest seminar, Oct. 27, 2015, Control Systems Group, Technical University of Berlin, Germany

l  Mental Workload Level Recognition by Combining Wavelet Packet Transform and Kernel Spectral Regression LDA Techniques, Guest seminar, Sept. 30, 2015, Control Systems Group, Technical University of Berlin, Germany

l  Multi-level recognition of momentary mental workload by using neurophysiologic feature reduction and ensemble support vector machine, Plenary lecture, 1st Arbeitstreffen zur Initiative “BioKybernetik”, Nov. 20-21, 2014, Grosskarlbach, Germany.

l  An Artificial Immune Algorithm based Selective Ensemble of Elman Neural Network Models for Forecast of Financial Time Series, Guest seminar, Nov. 17, 2014, Control Systems Group, Technical University of Berlin, Germany

l  Multi-level Recognition of Momentary Mental Workload Using Laplacian-Eigenmap-based Feature Reduction and Ensemble Support Vector Machines, Guest seminar, Nov. 14, 2014, Control Systems Group, Technical University of Berlin, Germany

l  Multi-level Recognition of Momentary Mental Workload Using Laplacian-Eigenmap-based Feature Reduction and Ensemble Support Vector Machines, Guest seminar, Nov. 11-12, 2014, Institute of Automation, Technical University of Dresden, Germany

l  Intelligent systems and control techniques with potential applications in aeronautic industry, Plenary lecture, Bilateral Workshop between ECUST and ACAE Commercial Aircraft Engine Co., Nov. 28, 2012, Shanghai, China

l  Intelligent model based prediction of human operator performance in complex and safety-critical human-machine systems: New results and implications for space flight, Plenary lecture, Bilateral Int. Conf. on Sustainable Futures between ECUST and The University of Nottingham, Nov. 05-06, 2012, Shanghai, China.

l  Pattern Classification of Human Mental Workload and Fatigue by Combining Recursive Feature Elimination and Support Vector Classification Techniques, Guest seminar, Oct. 29, 2012, Control Systems Group, Technical University of Berlin, Germany

l  Classifying Operator Functional State by Correlation Spectral Analysis and Hidden Markov Models, Guest seminar, Oct. 25-26, 2012, Institute of Automation, Technical University of Dresden, Germany

l  Classifying Operator Functional State by Correlation Spectral Analysis and Hidden Markov Models, Guest seminar, Oct. 24, 2012, Control Systems Group, Technical University of Berlin, Germany

l  Quantitative Assessment of Human Cognitive State: Fuzzy Modeling and Classification Approaches, Guest seminar, Nov. 09, 2011, Institute of Automation, Technical University of Dresden, Germany

l  Fuzzy Modeling and Classification of Human Operator Cognitive Functional State in Human-Machine Systems, Guest seminar, July 25, 2011, Control Systems Group, Technical University of Berlin, Germany

l  ECUST Intelligent Systems Research Group and A Proposal for Research Collaboration with TU Berlin, Guest seminar, Dec. 17, 2008, Control Systems Group, Technical University of Berlin, Germany

l  Beyond Conventional Automation: Intelligent Modeling, Analysis and Adaptive Automation of Complex and Safety-Critical Human-Machine Systems, Guest seminar, Dec. 17, 2008, Control Systems Group, Technical University of Berlin, Germany

l  Extracting cerebral evoked potentials: application of intelligent signal processing techniques, Guest seminar, Chair of Electrical Control Engineering (Dr F. Hoffmann), Universität Dortmund, Germany, Apr. 25, 2003.