**Structure of the book**

This book contains six applications of advanced analytics and AI in different industries. All the information is supported by practical examples and scientific detail. The chapters contain enough information for both beginners to become familiar with high technologies and science applications to solve business problems and advanced readers to acquire more detailed technical information.

In Chapter 1, an introductory review briefly gives a background to advanced analytics and AI applications to help industries make better decisions to optimize processes and reduce cost.

Chapter 2 is about using a bio-inspired hybrid algorithm for web services clustering. This chapter is written by researchers from the Autonomous Metropolitan University. In recent years, methods inspired by nature using biological analogies have been adapted for clustering problems, among which genetic algorithms, evolutionary strategies, and algorithms that imitate the behavior of some animal species have been implemented. In this chapter, researchers investigate how biologically inspired clustering methods can be applied to clustering web services and present a hybrid approach for web services clustering using the Artificial Bee Colony algorithm, K-Means, and Consensus. This hybrid algorithm was implemented and a series of experiments were conducted using three collections of web services. Results of the tests show that the solution approach is adequate and efficient to carry out the clustering of large groups of web services.

In Chapter 3, researchers from the Institute of Applied Economics, National Taiwan Ocean University, explain their achievements using optimization problem analysis for smart material planning. Mostly addressed is the concept of smart manufacturing, which is based on how to effectively facilitate production activities by using automation equipment; however, causing fluctuation in production may frequently root to the uncertain incoming sales orders. These uncertain factors may be determined by economic parameters, such as the changes of trading regulations and rivals' innovations, which require to be further deciphered to reduce risk and close the gap between forecasted and actual demand. This study presents a clear operable step-by-step framework to manage and cushion the impact of uncertain external factors. It also introduces three novel and feasible production planning models by considering the economic parameters.

Chapter 4 is related to the medical application of AI. In this chapter, a deep learning-based recommendation system for aesthetic surgery, composed of a mobile app and a deep learning model, has been proposed. Researchers from the Data Science Laboratory, FPT Software Japan Co., offer the deep learning model based on the dataset of before- and after-surgery facial images that can estimate the probability of the perfection of parts of a face. In this study, scientists focus on the two most popular treatments: rejuvenation treatment and eye double-fold surgery. In the project presented in this chapter, the researchers preliminarily achieved 88.9% and 93.1% accuracy on rejuvenation treatment and eye double-fold surgery, respectively.

Chapter 5 is written by a research group at the Department of Applied Mathematics in École Polytechnique. École Polytechnique is the only Grande École among all French universities and engineering schools with separate academic departments for pure and applied mathematics. The Department of Applied Mathematics focuses on the links between the many applications and all the major fields in science, engineering, and social sciences. The title of the chapter is "Assessment of the prediction quality of VPIN." VPIN is a tool designed to predict extreme events like flash crashes. Some concerns have been raised about its reliability. In this study, the researchers assess VPIN prediction quality (precision and recall rates) of extreme volatility events, including its sensitivity to the starting point of computation in each dataset. They benchmark the results with those of a "naive classifier." The test data used in this study contain five and a half years' worth of trading data of the five most liquid futures contracts of this period. The researchers found that VPIN has an unfortunate "flash crash" prediction power with the traditional 0.99 decision threshold. Increasing the decision threshold does not significantly improve overall prediction quality. Nevertheless, they found that VPIN has a more interesting predictive power for flash events of lower amplitude. Finally, the completed research showed that, for practice, the last bar price structure is the least sensitive to the starting point of computation.

Chapter 6 is about the application of AI in Earth observation. This chapter describes a Copernicus Access Platform Intermediate Layers Small-Scale Demonstrator, which is a comprehensive platform for the handling, analysis, and interpretation of Earth observation satellite images, mainly exploiting big data of the European Copernicus Program by AI methods. The main two components in Earth observation, namely data mining and data fusion, are detailed and validated in this study. The most important contributions of this chapter are the integration of these two components with a Copernicus platform on top of the European DIAS system for large-scale Earth observation image annotation, and the measurement of clustering and classification performances of various Copernicus Sentinel and third-party

**V**

mission data. This chapter is related to the completed research at the Remote

This book tries to give readers a better vision of advanced analytics and AI applications in different areas, and the authors hope that this volume will be a valuable resource for industry professionals and researchers. The presented chapters in this volume signify the state of the art regarding critical topics in advanced analytics and AI. The breadth of coverage and the depth in each of the sections make it a useful resource for all managers and engineers interested in the new generation of a data analytics applications. Above all, the editor hopes that this volume will spur on further discussions on all aspects of advanced analytics and AI applications in

**Ali Soofastaei**

Brisbane, Australia

Vale,

Artificial Intelligence Center,

Sensing Technology Institute, German Aerospace Center.

different industries.

mission data. This chapter is related to the completed research at the Remote Sensing Technology Institute, German Aerospace Center.

This book tries to give readers a better vision of advanced analytics and AI applications in different areas, and the authors hope that this volume will be a valuable resource for industry professionals and researchers. The presented chapters in this volume signify the state of the art regarding critical topics in advanced analytics and AI. The breadth of coverage and the depth in each of the sections make it a useful resource for all managers and engineers interested in the new generation of a data analytics applications. Above all, the editor hopes that this volume will spur on further discussions on all aspects of advanced analytics and AI applications in different industries.

> **Ali Soofastaei** Artificial Intelligence Center, Vale, Brisbane, Australia

**IV**

In Chapter 3, researchers from the Institute of Applied Economics, National Taiwan Ocean University, explain their achievements using optimization problem analysis for smart material planning. Mostly addressed is the concept of smart manufacturing, which is based on how to effectively facilitate production activities by using automation equipment; however, causing fluctuation in production may frequently root to the uncertain incoming sales orders. These uncertain factors may be determined by economic parameters, such as the changes of trading regulations and rivals' innovations, which require to be further deciphered to reduce risk and close the gap between forecasted and actual demand. This study presents a clear operable step-by-step framework to manage and cushion the impact of uncertain external factors. It also introduces three novel and feasible production planning models by

Chapter 4 is related to the medical application of AI. In this chapter, a deep learning-based recommendation system for aesthetic surgery, composed of a mobile app and a deep learning model, has been proposed. Researchers from the Data Science Laboratory, FPT Software Japan Co., offer the deep learning model based on the dataset of before- and after-surgery facial images that can estimate the probability of the perfection of parts of a face. In this study, scientists focus on the two most popular treatments: rejuvenation treatment and eye double-fold surgery. In the project presented in this chapter, the researchers preliminarily achieved 88.9% and 93.1% accuracy on rejuvenation treatment and eye double-fold surgery, respectively.

Chapter 5 is written by a research group at the Department of Applied Mathematics in École Polytechnique. École Polytechnique is the only Grande École among all French universities and engineering schools with separate academic departments for pure and applied mathematics. The Department of Applied Mathematics focuses on the links between the many applications and all the major fields in science, engineering, and social sciences. The title of the chapter is "Assessment of the prediction quality of VPIN." VPIN is a tool designed to predict extreme events like flash crashes. Some concerns have been raised about its reliability. In this study, the researchers assess VPIN prediction quality (precision and recall rates) of extreme volatility events, including its sensitivity to the starting point of computation in each dataset. They benchmark the results with those of a "naive classifier." The test data used in this study contain five and a half years' worth of trading data of the five most liquid futures contracts of this period. The researchers found that VPIN has an unfortunate "flash crash" prediction power with the traditional 0.99 decision threshold. Increasing the decision threshold does not significantly improve overall prediction quality. Nevertheless, they found that VPIN has a more interesting predictive power for flash events of lower amplitude. Finally, the completed research showed that, for practice, the last bar price structure is the least sensitive to the

Chapter 6 is about the application of AI in Earth observation. This chapter describes a Copernicus Access Platform Intermediate Layers Small-Scale Demonstrator, which is a comprehensive platform for the handling, analysis, and interpretation of Earth observation satellite images, mainly exploiting big data of the European Copernicus Program by AI methods. The main two components in Earth observation, namely data mining and data fusion, are detailed and validated in this study. The most important contributions of this chapter are the integration of these two components with a Copernicus platform on top of the European DIAS system for large-scale Earth observation image annotation, and the measurement of clustering and classification performances of various Copernicus Sentinel and third-party

considering the economic parameters.

starting point of computation.

**1**

from hidden data trends [1].

**Chapter 1**

*Ali Soofastaei*

**1. Introduction**

Introductory Chapter: Advanced

*"The key challenge is not so much globalization. It is what I call the fourth industrial revolution. Because its technology which creates major changes in our daily lives. It's a technology that creates fears. What we want to do is make the world much more aware. On the one hand of the opportunity of the new technology but* 

 *—Klaus Schwab*

The opportunities and complexities associated with the digital era can be overwhelming to industries and markets, which face an enormous amount of potential information in each transaction. Being aware of trends in the data pool and benefiting from hidden information has created a new paradigm, redefining the meaning of corporate power. Access to information can make organizations more effective and help them to reach their goals. Big data analytics (BDA) enables industries to describe, diagnose, predict, prescribe, and find hidden growth opportunities, potentially increasing business value. BDA uses advanced analytical techniques to enhance knowledge and improve decision-making by reducing the complexity of exponentially increasing amounts of data. BDA uses novel and sophisticated algorithms to analyze real-time data, resulting in highly accurate analytics. Depending on the problem being solved, these complex algorithms can be allocated to either

A significant consequence of the digital world is the creation of bulk raw data.

Managers are responsible for managing this valuable capital, with its various shapes and sizes, on the basis of organizational needs. Big data has the power to affect all aspects of society, from social to educational. As the volume of raw data increases, particularly in technology-based companies, the issue of managing it becomes more critical. The variety, velocity, and volume of raw data warrant the use of advanced tools to overcome its complexity and to reveal the hidden information embedded in it. Thus, BDA has been proposed as a means of experimentation, simulation, data analysis, and monitoring. One BDA tool, advanced analytics (AA), can provide the foundation for predictive analysis on the basis of supervised and unsupervised data input. A reciprocal relationship exists between the power of AA and data input—the more precise and accurate the input data, the more effective the analytical performance. Additionally, ML, artificial intelligence (AI) and deep learning as subfields of AA can be used to extract knowledge

Analytics and Artificial

Intelligence Applications

*on the other hand the risks and dangers we encounter".* 

deep learning or machine learning (ML) approaches.
