| Description |
1 online resource (246 p.). |
| Series |
Issn Ser. |
|
Issn Ser.
|
| Note |
Description based upon print version of record. |
| Contents |
Intro -- AI and Cloud Computing -- Copyright -- Contents -- Contributors -- Preface -- Chapter One: A Deep-forest based approach for detecting fraudulent online transaction -- 1. Introduction -- 2. What is deep forest -- 2.1. Basics of deep forest -- 2.1.1. Decision tree -- 2.1.2. Random Forest -- 2.2. Cascade forest structure -- 2.3. Multi-grained scanning -- 3. Machine learning for detecting fraud online transaction -- 3.1. Definition of supervised learning -- 3.2. Classification task -- 3.3. Supervised learning for fraud detection -- 3.3.1. Supervised profiling method |
|
3.3.2. Classification method -- 3.3.3. Cost-sensitive method -- 3.3.4. Network methods -- 3.4. Unsupervised learning for fraud detection -- 3.5. Feature engineering for fraud detection -- 4. Pressing issues in detecting fraud online transaction -- 4.1. Sampling methods -- 4.2. Cost-based methods -- 5. Case study: Deep forest in detecting online transaction fraud -- 5.1. GcForest-based fraud transaction detection framework -- 5.2. Transaction time-based differentiation feature generation method -- 5.3. GcForest-based model with outlier samples detection -- 5.4. Experiment and evaluation |
|
5.5. Future research -- 6. Summary of key lessons learned -- 7. Conclusion -- Acknowledgments -- Key terminology and definitions -- References -- Chapter Two: Design of cyber-physical-social systems with forensic-awareness based on deep learning -- 1. Introduction -- 2. Issues in discussion -- 2.1. Threats of image forgery -- 2.2. Image forensic methods -- 2.3. Major type forgery detections -- 2.3.1. Clone detection -- 2.3.2. Splicing detection -- 3. Forgery detection with deep learning -- 3.1. Convolutional Neural Network in image classification -- 3.2. Literature review |
|
3.3. Generation of a multichannel feature map -- 3.3.1. Hidden periodicity of interpolated signals -- 3.4. Periodicity property of forensic -- 3.5. Filtering residual feature map -- 4. Our scheme: Toward a universal system of forgery detection -- 4.1. Micro neural networks -- 4.2. Generation of f-map layer -- 4.3. Convolutional module -- 4.4. Classification module -- 5. Experiment results -- 5.1. Experiment setup -- 5.2. Experimental results -- 5.3. Robustness against JPEG attack -- 6. Summary of key contribution of the research -- 7. Research directions in the field -- 8. Conclusion |
|
Acknowledgments -- Key terminology and definitions -- References -- Chapter Three: Review on privacy-preserving data comparison protocols in cloud computing -- 1. Introduction -- 2. Issues in discussion -- 3. Comparison protocols for curious data owners (CDO) -- 3.1. Framework of comparison protocols for CDO -- 3.2. Protocols with general public key encryption -- 3.3. Protocols with homomorphic encryption -- 3.3.1. Sign bit decryption method -- 3.3.2. Difference comparison method -- 3.4. Analysis of comparison protocols for CDO -- 3.4.1. The security of the protocols |
| Note |
3.4.2. The performance of the protocols. |
| Added Author |
Wu, Sheng.
|
| Other Form: |
Print version: Hurson, Ali R. AI and Cloud Computing San Diego : Elsevier Science & Technology,c2021 9780128211472 |
| ISBN |
9780128211489 |
|
0128211482 |
|
