XAI Based Intelligent Systems for Society 5.0 / edited by Fadi Al-Turjman [and more].

Format:

Book

Contributor:

Al-Turjman, Fadi.

Nayyar, Anand.

Naved, Mohd.

Singh, Anuj K.

Vēlāyutan̲, T. A.

ProQuest ebook central

Language:

English

Subjects (All):

Artificial intelligence--Industrial applications.

Artificial intelligence.

Cooperating objects (Computer systems).

Physical Description:

1 online resource (428 p.)

Place of Publication:

Amsterdam, Netherlands ; Oxford, United Kingdom ; Cambridge MA : Elsevier, [2024]

Contents:

Front Cover

XAI Based Intelligent Systems for Society 5.0

Copyright

Contents

List of contributors

I

Paradigm shift and history of XAI

1

Paradigm shift from AI to XAI of Society 5.0: Machine-centric to human-centric

1. Introduction

1.1 Concept of artificial intelligence

1.2 Evolution of artificial intelligence and IoT in business

1.3 Impact of artificial intelligence on business functions

1.3.1 Artificial intelligence in manufacturing

1.3.2 Artificial intelligence in human resource management

1.3.3 Artificial intelligence in marketing

1.3.4 Artificial intelligence in finance

1.4 Ethical consideration of AI

1.4.1 Organization of chapter

2. Concept of XAI of Society 5.0

2.1 Evolution of XAI of Society 5.0

2.2 Four principles of explainable AI

2.2.1 Principle of explanation

2.2.2 Principle of meaningful

2.2.3 Principle of explanation accuracy

2.2.4 Principle of knowledge limits

3. Shift from machine-centric to human-centric

3.1 Applicability in real life and challenges of eXplainable artificial intelligence

3.1.1 Real-time applications

3.1.2 Real ethical concerns around XAI

3.2 Future of XAI

4. Conclusion and future scope

4.1 Conclusion

References

2

Towards explainable artificial intelligence: history, present scenarios, and future trends

1.1 Objectives of the chapter

1.2 Organization of the chapter

2. From AI to XAI

2.1 Artificial Intelligence

2.2 Explainable Artificial Intelligence

3. The what, why and how of XAI

3.1 The what

3.2 The why

3.2.1 Difficulty in explaining or modeling interpretability

3.2.2 Trust

3.2.3 Explanations are part of legislation

3.2.4 Bias and transparency

3.2.5 Fairness

3.2.6 Scientific usage and explanations are a prerequisite for new insights

3.2.7 Ethical and legal requirements

3.2.8 Human limitation

3.3 The how

3.3.1 Methods of XAI

3.3.1.1 Numerical explanation

3.3.1.2 Rule-based explanation

3.3.1.3 Visual explanation

3.3.1.4 Mixed explanation

3.3.2 Approaches of XAI

3.3.2.1 Local interpretable model-agnostic explanations

3.3.2.2 SHapley Additive exPlanation (SHAP)

3.3.2.3 GRADient Class Activation Mapping (GRAD-CAM)

3.3.2.4 Deep Learning Important FeaTures (DeepLIFT)

4. Recent development trends in XAI

4.1 Regulations

4.2 Privacy protection

4.3 Human-centered approaches to XAI

4.4 Evaluation of explanation methods

4.5 Commercial barrier

4.6 Hybridization

5. Perspective toward what is yet to be achieved in the field

5.1 User experience

5.2 Standardization

5.3 The balance between explanation and performance

5.4 Evaluation

5.5 Ground truths

5.6 Research collaboration

5.7 Methods associated with explanation

5.7.1 Security

5.7.2 Types of explanation and timeliness

5.7.3 Explainable methods

5.7.4 Uncertainties

Notes:

Includes bibliographical references and index.

5.7.5 Cost

Electronic reproduction. Ann Arbor, MI Available via World Wide Web.

Description based on online resource; title from digital title page (viewed on February 13, 2024).

Other Format:

Print version: Al-Turjman, Fadi XAI Based Intelligent Systems for Society 5. 0

ISBN:

0323957846

9780323957847

Publisher Number:

40032143928

Access Restriction:

Restricted for use by site license.