Artificial Neural Networks in Pattern Recognition synthesizes the proceedings of the 4th IAPR TC3 Workshop, ANNPR 2010. Topics include supervised and unsupervised learning, feature selection, pattern recognition in signal and image processing.

The 4th IAPR TC3 Workshop on Arti?cial Neural Networks in Pattern Rec- nition, ANNPR 2010, was held at Nile University (Egypt), April 11 13, 2010. The workshop was organized by the Technical Committee on Neural Networks and Computational Intelligence (TC3) that is one of the 20 technical comm- tees (TC) of the International Association for Pattern Recognition (IAPR). The scope of TC3 includes computational intelligence approaches, such as fuzzy s- tems, evolutionary computing and arti?cial neural networks and their use in various pattern recognition applications. The major topics of ANNPR are supervised and unsupervised learning, f- ture selection, pattern recognition in signal and image processing, and appli- tions in data mining or bioinformatics. High quality across such a diverse ?eld of research is achieved through a rigorous and selective review process. For this workshop, 42 papers were submitted and 23 of them were selected for inc- sion in the proceedings. The workshop was enriched by three invited talks given by Barbara Hammer, University of Bielefeld, Germany, Amir F. Atiya, Cairo University, Egypt, and Mohamed Kamel, University of Waterloo, Canada. We would like to thank all authors for the e?ort they put into their subm- sions, and the Scienti?c Committee for taking the time to provide high-quality reviews and selecting the best contributions for the ?nal workshop program. Special thanks are due to the members of the Nile University Organizing C- mittee,AhmedSalah,AmiraElBaroudy,EsraaAly,HebaEzzat,NesrineSameh, Rana Salah and Mohamed Zahhar for their indispensable contributions to the registration management and local organization.

Fast track conference proceeding Unique visibility State of the art research

Zusammenfassung
Artificial Neural Networks in Pattern Recognition synthesizes the proceedings of the 4th IAPR TC3 Workshop, ANNPR 2010. Topics include supervised and unsupervised learning, feature selection, pattern recognition in signal and image processing.

Inhalt
Supervised Learning.- Pattern Classification Using a Penalized Likelihood Method.- Evaluation of Feature Selection by Multiclass Kernel Discriminant Analysis.- Correlation-Based and Causal Feature Selection Analysis for Ensemble Classifiers.- A New Monte Carlo-Based Error Rate Estimator.- Recognition of Sequences of Graphical Patterns.- Maximum Echo-State-Likelihood Networks for Emotion Recognition.- Robustness Analysis of Eleven Linear Classifiers in Extremely HighDimensional Feature Spaces.- Unsupervised Learning.- Global Coordination Based on Matrix Neural Gas for Dynamic Texture Synthesis.- SIC-Means: A Semi-fuzzy Approach for Clustering Data Streams Using C-Means.- The Mathematics of Divergence Based Online Learning in Vector Quantization.- Cluster Analysis of Cortical Pyramidal Neurons Using SOM.- Parallelized Kernel Patch Clustering.- Visual Pattern Recognition.- Neural Network Cascade for Facial Feature Localization.- A Hidden Markov Model Based Approach for Facial ExpressionRecognition in Image Sequences.- Analysis, Interpretation, and Recognition of Facial Action Units and Expressions Using Neuro-Fuzzy Modeling.- Content-Based Retrieval and Classification of Ultrasound Medical Images of Ovarian Cysts.- Applications.- A Novel Word Spotting Algorithm Using Bidirectional Long Short-Term Memory Neural Networks.- Swarm Based Fuzzy Discriminant Analysis for Multifunction Prosthesis Control.- Bayesian Learning of Generalized Gaussian Mixture Models on Biomedical Images.- Defective Areas Identification in Aircraft Components by Bivariate EMD Analysis of Ultrasound Signals.- Different Regions Identification in Composite Strain-Encoded (C-SENC) Images Using Machine Learning Techniques.- Exploiting Neural Networks to Enhance Trend Forecasting for Hotels Reservations.- VLSI Architecture of the Fuzzy Fingerprint Vault System.- Invited Talk.- Clustering Very Large Dissimilarity Data Sets.