| Description |
1 online resource (358 p.) |
| Series |
Smart Innovation, Systems and Technologies ; v.360 |
|
Smart innovation, systems, and technologies ; 360.
|
| Contents |
Intro -- Program Committee -- Sponsoring Institutions -- Preface -- Contents -- About the Editors -- Part I Neural Networks and Related Applications -- 1 Generating New Sounds by Vector Arithmetic in the Latent Space of the MelGAN Architecture -- 1.1 Introduction -- 1.2 Proposed Idea -- 1.2.1 GAN: Generative Adversarial Network -- 1.2.2 The Used MelGAN -- 1.2.3 The Vector Arithmetic in the Latent Space -- 1.3 Experimental Setup -- 1.4 Simulation Results -- 1.5 Conclusion -- References -- 2 Graph Neural Networks for Topological Feature Extraction in ECG Classification -- 2.1 Introduction |
|
2.2 Methodology -- 2.2.1 From Time Series to Complex Network -- 2.2.2 Graph Isomorphism Network -- 2.3 Experiment -- 2.4 Conclusion -- References -- 3 Manifold Learning by a Deep Gaussian Process Variational Autoencoder -- 3.1 Introduction -- 3.2 Deep Gaussian Processes -- 3.3 Deep Gaussian Processes Variational Autoencoder -- 3.4 Experimental Validation -- 3.5 Conclusions -- References -- 4 Analysis of Sensors for Movement Analysis -- 4.1 Introduction -- 4.2 Hand Movement Analysis -- 4.2.1 Microchip DV102014 -- 4.2.2 Leap Motion -- 4.2.3 Noitom Perception Mocap |
|
4.2.4 Android App for Tapping Analysis -- 4.3 Experimental Results of Hand Movement Analysis -- 4.3.1 Experimental Setup for Leap Motion and Noitom Mocap -- 4.3.2 Static Analysis Results -- 4.3.3 Dynamic Analysis Results -- 4.4 Foot Sensor Design and Foot Movement Analysis -- 4.5 Experimental Results of Foot Movement Analysis -- 4.5.1 Experimental Setup -- 4.5.2 Results -- 4.6 Conclusions -- References -- 5 Dual Deep Clustering -- 5.1 Introduction -- 5.2 Gradient-Based Competitive Learning -- 5.2.1 Duality Theory for Single-Layer Networks |
|
5.2.2 Loss Minimization for Clustering and Topological Analysis -- 5.3 Simulations -- 5.3.1 High-Dimensional Simulations -- 5.4 Theoretical Interpretation -- 5.5 Conclusions -- References -- 6 Learning-Based Approach to Predict Fatal Events in Brugada Syndrome -- 6.1 Brief Introduction to the Brugada Syndrome -- 6.1.1 Risk Stratification -- 6.2 Database Description -- 6.3 Learning Architectures -- 6.4 BDT and MLP Experiments -- 6.4.1 Supplementary Experiments -- 6.5 Discussion -- References -- 7 Breast Cancer Localization and Classification in Mammograms Using YoloV5 -- 7.1 Introduction |
|
7.2 Related Work -- 7.3 Materials and Methods -- 7.3.1 Datasets -- 7.3.2 Data Pre-processing -- 7.3.3 YoloV5 -- 7.3.4 Data Augmentation -- 7.4 Results -- 7.5 Discussion and Conclusion -- References -- 8 Deep Acoustic Emission Detection Trained on Seismic Signals -- 8.1 Introduction -- 8.2 Methodology -- 8.3 Results and Discussion -- 8.4 Conclusions and Future Developments -- References -- 9 A Deep Learning Framework for the Classification of Pre-prodromal and Prodromal Alzheimer's Disease Using Resting-State EEG Signals -- 9.1 Introduction -- 9.2 Related Work -- 9.3 Materials and Methods |
| Summary |
This book provides an overview on the current progresses in artificial intelligence and neural nets in data science. The book is reporting on intelligent algorithms and applications modeling, prediction, and recognition tasks and many other application areas supporting complex multimodal systems to enhance and improve humanmachine or humanhuman interactions. This field is broadly addressed by the scientific communities and has a strong commercial impact since investigates on the theoretical frameworks supporting the implementation of sophisticated computational intelligence tools. Such tools will support multidisciplinary aspects of data mining and data processing characterizing appropriate system reactions to human-machine interactional exchanges in interactive scenarios. The emotional issue has recently gained increasing attention for such complex systems due to its relevance in helping in the most common human tasks (like cognitive processes, perception, learning, communication, and even "rational" decision-making) and therefore improving the quality of life of the end users |
| Notes |
9.4 Results |
| Subject |
Data mining.
|
|
Artificial intelligence -- Data processing.
|
|
Neural networks (Computer science)
|
|
Artificial intelligence -- Data processing.
|
|
Data mining.
|
|
Neural networks (Computer science)
|
| Form |
Electronic book
|
| Author |
Esposito, Anna
|
|
Faúndez Zanuy, Marcos.
|
|
Morabito, F. C. (Francesco Carlo)
|
|
Pasero, Eros
|
| ISBN |
9789819935925 |
|
981993592X |
|
