%{search_type} search results

• Intro; Preface; Organization; Contents
• Part I; Contents
• Part II; Neural Networks and Their Applications; Author Profiling with Classification Restricted Boltzmann Machines; 1 Introduction; 2 Author Profile Dimensions; 3 Restricted Boltzmann Machines; 4 Probabilities and Gradients; 4.1 Discriminative Training; 4.2 Generative Training; 5 Evaluation Datasets; 6 Experiments and Results; 6.1 Overall Results; 7 Conclusions; References; Parallel Implementation of the Givens Rotations in the Neural Network Learning Algorithm; 1 Introduction; 2 Givens Elimination Step; 3 Givens QR Decomposition.

• 4 QR Decomposition in Neural Network Weights Update5 Parallel Implementation; 6 Simulation Results; 7 Conclusion; References; Parallel Levenberg-Marquardt Algorithm Without Error Backpropagation; 1 Introduction; 2 Parallel Realisation; 2.1 Calculating the Weight Derivatives Without Error Backpropagation; 2.2 Calculating the A Matrix and the Gradient Vector; 2.3 The QR Decomposition Based on the Householser Reflections; 3 Computational Results; 4 Conclusions; References; Spectral Analysis of CNN for Tomato Disease Identification; Abstract; 1 Introduction; 2 Related Works.

• 3 Spectral Analysis of CNN for Tomato Disease3.1 Deep Visualization of CNN; 3.2 Color Sensitivity of RGB Images; 3.3 Sensitivity to Color with Different Wavelength Values; 3.3.1 Visible Spectrum of Images; 4 Experimental Results; 4.1 Dataset Description; 4.2 CNN Activations and Features Visualization; 4.2.1 Activations of Neurons; 4.2.2 RGB Color Sensitivity; 4.2.3 Feature Maps; 5 Conclusion and Future Work; Acknowledgments; References; From Homogeneous Network to Neural Nets with Fractional Derivative Mechanism; 1 Introduction; 2 Weight Distribution with Fractional Calculus.

• 3 Fractional Derivative Inside Neuron Transfer Function4 The Fractional Mechanism Within 2D Homogeneous Network; 5 Conclusion; References; Neurons Can Sort Data Efficiently; 1 Introduction; 2 Models of Neurons, Receptors, and the Senses; 2.1 Sensory Fields and Sensors; 2.2 Extreme, Sensory and Object Neurons; 3 Simplistic Sequential Neural Associative Sorting; 4 Conclusions and Remarks; References; Avoiding Over-Detection: Towards Combined Object Detection and Counting; 1 Introduction; 2 Related Work; 2.1 Deep Learning Methods for Object Detection; 2.2 Deep Learning Methods for Cell Detection.

• 3 Method3.1 Loss Function; 3.2 Model Architecture; 4 Results; 5 Conclusion; References; Echo State Networks Simulation of SIR Distributed Control; 1 Introduction; 2 Echo State Networks; 3 SIR Model with Delay and Spatial Diffusions; 3.1 Distributed Optimal Control Problem; 4 Discretisation and Adaptive Critic Neural Networks Solution of the Distributed Optimal Control; 4.1 Numerical Simulation; 5 Conclusion; References; The Study of Architecture MLP with Linear Neurons in Order to Eliminate the ``vanishing Gradient'' Problem; 1 Introduction; 2 Nonlinearity capabilities of deep neural networks.

The two-volume set LNAI 10245 and LNAI 10246 constitutes the refereed proceedings of the 16th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2017, held in Zakopane, Poland in June 2017. The 133 revised full papers presented were carefully reviewed and selected from 274 submissions. The papers included in the second volume are organized in the following five parts: data mining; artificial intelligence in modeling, simulation and control; various problems of artificial intelligence; special session: advances in single-objective continuous parameter optimization with nature-inspired algorithms; special session: stream data mining.
(source: Nielsen Book Data)
The two-volume set LNAI 10245 and LNAI 10246 constitutes the refereed proceedings of the 16th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2017, held in Zakopane, Poland in June 2017. The 133 revised full papers presented were carefully reviewed and selected from 274 submissions. The papers included in the first volume are organized in the following five parts: neural networks and their applications; fuzzy systems and their applications; evolutionary algorithms and their applications; computer vision, image and speech analysis; and bioinformatics, biometrics and medical applications.
(source: Nielsen Book Data)

• Intro; Preface; Organization; Contents
• Part II; Contents
• Part I; Data Mining; Computer Based Stylometric Analysis of Texts in Polish Language; 1 Introduction; 2 Features Used in Stylometry; 3 Performed Stylometric Analysis; 3.1 Data Sets and Classes; 3.2 Feature Generation Methods; 3.3 Classification; 4 Results; 4.1 Stylometric Features; 4.2 Training Set Size; 4.3 Classifiers Accuracy; 4.4 Statistical Features Importance; 4.5 Grammatical Classes Importance; 5 Summary; References; Integration Base Classifiers Based on Their Decision Boundary; 1 Introduction; 2 Basic Concept.

• 3 Proposed Method4 Experimental Studies; 5 Conclusion; References; Complexity of Rule Sets Induced by Two Versions of the MLEM2 Rule Induction Algorithm; 1 Introduction; 2 Incomplete Data Sets; 3 Probabilistic Approximations; 4 Rule Induction; 4.1 True MLEM2; 4.2 Emulated MLEM2; 5 Experiments; 6 Conclusions; References; Spark-Based Cluster Implementation of a Bug Report Assignment Recommender System; 1 Introduction; 2 Related Work; 3 The Recommender System; 3.1 Datasets; 3.2 Cleansing; 3.3 Preprocessing and Feature Reduction; 3.4 Training the Recommender; 3.5 Implementation.

• 4 Results and Discussion4.1 Results of Dimensionality Reduction Techniques; 4.2 Choice of SVM Kernel; 4.3 Comparison with Previous Approaches and Scalability; 5 Conclusion; References; The Bag-of-Words Method with Dictionary Analysis by Evolutionary Algorithm; 1 Introduction; 2 Description of Proposed Methods; 3 Experimental Research; 4 Conclusions; References; The Novel Method of the Estimation of the Fourier Transform Based on Noisy Measurements; 1 Introduction; 2 Orthogonal Series Estimation of Regression Function and Its Spectrum.

• 3 Nonparametric Orthogonal Series Estimation of Fourier Transform4 Simulation Example; 5 Remarks and Extensions; References; A Complete Efficient FFT-Based Algorithm for Nonparametric Kernel Density Estimation; 1 Introduction; 2 Problem Demonstration; 3 The Improved FFT-Based Algorithm for Density Estimation; 4 Binning; 5 Experiments; 6 Conclusions; References; A Framework for Business Failure Prediction; 1 Introduction; 2 Related Work; 3 The Dataset and the Proposed Framework; 3.1 Details of Dataset; 3.2 Proposed Framework; 4 Performance Evaluation Results; 5 Conclusions; References.

• Fuzzy Clustering with -Hyperballs and Its Application to Data Classification1 Introduction; 2 Fuzzy Clustering with -Hyperballs; 3 FCH Based Classification; 4 Results and Discussion; 5 Conclusions; References; Two Modifications of Yinyang K-means Algorithm; 1 Introduction; 2 Yinyang K-means; 3 Modifications of the Algorithm; 3.1 Regrouping of Centroids; 3.2 Initial Grouping and Regrouping Using Same-Size K-means; 4 Experimental Results; 5 Conclusions; References; Detection of the Innovative Logotypes on the Web Pages; 1 Introduction; 2 Related Works vs. Proposed Approach.

The two-volume set LNAI 10245 and LNAI 10246 constitutes the refereed proceedings of the 16th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2017, held in Zakopane, Poland in June 2017. The 133 revised full papers presented were carefully reviewed and selected from 274 submissions. The papers included in the second volume are organized in the following five parts: data mining; artificial intelligence in modeling, simulation and control; various problems of artificial intelligence; special session: advances in single-objective continuous parameter optimization with nature-inspired algorithms; special session: stream data mining.
(source: Nielsen Book Data)