Title: Advances in Artificial Intelligence

  Editor: Sergey Y. Yurish

  Publisher: International Frequency Sensor Association (IFSA) Publishing, S. L.

  Formats: pdf Acrobat (e-book), 120 pages

  Pubdate: 10 February 2023

  e-ISBN: 978-84-09-47562-9

Open Access book in pdf format

(free download, 5.69 MB):

 Book Description

Artificial intelligence has is one of the fastest-growing technologies in recent years. The market growth is mainly driven by factors such as the increasing adoption of cloud-based applications and services, growing big data, and increasing demand for intelligent virtual assistants. Various end-use industries have also employed artificial intelligence such as retail and business analysis that has also boosted the demand in this market. The major restraint for the market is the limited number of artificial intelligence technology experts. The Book Series on ‘Advances in Artificial Intelligence’ has been launched with the aim to fill-in this gap.

The book volume contains four chapters written by 13 contributors from three countries: France, Iran and Slovenia.

All chapters have the same structure: first an introduction to specific topic under study; second particular field description including appropriate applications. Each of chapter is ending by well selected list of references with books, journals, conference proceedings and web sites.

This book ensures that our readers will stay at the cutting edge of the field and get the right and effective start point and road map for the further researches and developments.

With this unique combination of information in each volume, the ‘Advances in Artificial Intelligence’ book Series will be of value for scientists and engineers in industry and at universities, to developers and users.

Contents:

Contents
Preface
Contributors

 
1. Fuzzy-based Optimisation of Unit Selection Algorithm for Corpus-based TTS Systems

1.1. Introduction
1.2. The Unit Selection Algorithm in Corpus-based TTS Systems
1.3. A Lossy and Non-lossy Compression  of Concatenation Costs
1.4. Fuzzy-based Unit Selection Algorithm Optimization	
1.5. Discussion
1.6. Conclusion
Acknowledgements
References

2. Graphs as Tools to Improve Deep Learning Methods

2.1. Introduction
2.2. Background
2.2.1. Deep Neural Networks for Computer Vision
2.2.2. Graph Signal Processing
2.3. Graphs as Tools to Improve Deep Learning Methods
2.3.1. Graphs to Interpret Intermediate Representations
2.3.2. Graphs to Denoise Intermediate Representations
2.3.3. Graphs as Losses
2.3.4. Graphs as Regularizers
2.4. Conclusion
References

3. Machine Learning in Peer-to-peer Lending – The New Theme Park of Financial Risk Modelling

3.1. Introduction
3.2. Methodology
3.2.1. Data
3.2.2. Definition of Prediction Model
3.2.3. Validation Procedure
3.3. Results
3.3.1. Parameter Sensitivity
3.3.2. Number of Terms in Function
3.3.3. Validation of Prediction Model
3.3.4. Comparison with Machine Learning Algorithms
3.3.5. Interpretation of the Prediction Model
3.4. Discussion
Acknowledgements
References

4. Early Detection of Mild Alzheimer’s Diseases with Combine MRI  and EEG Signals

4.1. State the Problem
4.2. Necessity of Doing Research
4.3. Methods of Diagnosing Alzheimer's disease
4.4. EEG Signal Preprocessing
4.5. Power Spectrum
4.6. Time-frequency Analysis
4.6.1. Time-frequency Analysis with Short-term Instantaneous Conversion
4.7. Wavelet Properties
4.8. Function and Correlation Coefficient
4.9. Coherency
4.10. Nonlinear Features
4.11. Lyapunov's Exponent
4.12. Correlation Dimension
4.13. Entropy
4.14. Optimal Feature Selection
4.15. Classification
4.16. Methods for Dividing Data into Training and Test Categories and Validation Methods
4.17. Labeling Steps
4.18. Type and Number of Channels to Record  the Signal
4.19. Signal Recording Protocol
4.20. Research Volunteers
4.21. Brain Electrical Signals
4.22. P300
4.23. Origin P300
4.24. Oddball Pattern for P300 Extraction
4.25. Results of Two Neural Networks, Canalization  and Perceptron Network
4.26. Discussion and Conclusion
References

Index

To be continued.