Preface

Machine learning (ML) is the ability of a system to automatically acquire, integrate, and then develop knowledge from large-scale data, and then expand the acquired knowledge autonomously by discovering new information, without being specifically programmed to do so. In short, the ML algorithms can find application in the following: (1) a deeper understanding of the cyber event that generated the data under study, (2) capturing the understating of the event in the form of a model, (3) predicting the future values that will be generated by the event based on the constructed model, and (4) proactively detect any anomalous behavior of the phenomenon so that appropriate corrective actions can be taken beforehand. ML is an evolutionary area, and with recent innovations in technology, especially with the development of smarter algorithms and advances in hardware and storage systems, it has become possible to perform a large number of tasks more efficiently and precisely, which were not even imaginable a couple of decades before. Over the past few years, deep learning (DL), a specialized subset of machine learning that involves more complex architectures, algorithms, and models for solving complex problems and predicting future outcomes of complex events, has also evolved.

Recent times are witnessing rapid development in machine learning algorithm systems, especially in reinforcement learning, natural language processing, computer and robot vision, image processing, speech, and emotional processing and understanding. There are numerous applications of machine learning that have emerged or are evolving at present in several business domains, such as medicines and healthcare, finance and investment, sales and marketing, operations and supply chain, human resources, media and entertainment, and so on.

Of late, the applied ML systems in the industry are exhibiting some prominent trends. These trends will utilize the power of ML and artificial intelligence (AI) systems even further to reap benefits in business and society, in general. Some of these trends are as follows: (1) less volume of code and faster implementation of ML systems; (2) increasing use of light-weight systems suitable for working on the resource-constrained internet of things (IoT) devices; (3) automatic generation of codes for building ML models; (4) designing novel processes for robust management of the development of ML systems for increased reliability and efficiency; (5) more wide-spread adoption of deep-learning solutions into products of all domains and applications; (6) increased use of generative adversarial networks (GAN)-based models for various image processing applications including image searching, image enhancement, etc.; (7) more prominence of unsupervised learning-based systems that require less or no human intervention for their operations; (8) use of reinforcement learning-based systems; and finally, (9) evolution of few-shots, if not zero-shot learning-based systems.

In tune with the increasing importance and relevance of ML models, algorithms, and their applications and with the emergence of more innovative uses-cased of DL- and AI-based systems, the current volume presents a few innovative research works

and their applications in real-world, such as stock trading, medical and healthcare systems, and software automation. The chapters in the book illustrate how ML and DL algorithms and models are designed, optimized, and applied for achieving higher precision and efficiency in business and other processes in real-world scenarios.

The book presents six chapters that highlight different architectures, models, algorithms, and applications of machine learning, deep learning, and artificial intelligence. The subject matter discussed in the chapters of the book illustrates the complexities involved in the design, training, validation, testing, and deployment of machine learning and deep learning models in real-world applications.

In the introductory chapter entitled *Introductory Chapter: Machine Learning in Finance-Emerging Trends and Challenges*, Jaydip Sen, Rajdeep Sen, and Abhishek Dutta present a landscape of the emerging trends and challenges in banks and other financial institutions that machine learning-based models will face in the near future and how the organizations should transform those challenges into opportunities for their business.

In Chapter 2, *Design and Analysis of Robust Deep Learning Models for Stock Price Prediction*, Jaydip Sen and Sidra Mehtab present a set of deep learning-based regression models for precise and robust prediction of future prices of a stock listed in the diversified sector in the National Stock Exchange (NSE) of India. The models are built on convolutional neural network (CNN) and long short-term memory (LSTM) network architectures, and they are designed to handle high-frequency stock price data. The performances of the models are compared based on their execution time and prediction accuracies.

In Chapter 3, *Articulated Human Pose Estimation Using Greedy Approach*, Pooja Kherwa, Sonali Singh, Saheel Ahmed, Prannay Berry, and Sahil Khurana propose a method of bottom-up parsing for human pose estimation that localizes anatomical key points of humans. The authors have also presented results on the performance of their proposed mechanism.

In Chapter 4, *Ensemble Machine Learning Algorithms for Prediction and Classification of Medical Images*, Racheal S. Akinbo and Oladunni A. Daramola discuss how ensemble machine learning models can be applied for classifying medical images, such as ultrasound reports, X-rays, mammography, fluoroscopy, computer-aided tomography, magnetic resonance image, magnetic resonance angiography, nuclear medicine, and positron emission tomography.

In Chapter 5, *Delivering Precision Medicine to Patients with Spinal Cord Disorders; Insights into Applications of Bioinformatics and Machine Learning from Studies of Degenerative Cervical Myelopathy*, Kalum J. Ost, David W. Anderson and David W. Cadotte present an approach towards designing a machine learning-based clinical diagnosis system that supports the diagnostic prediction of DCM severity. The proposed system is intended to serve as a support system to the patients rather than a recommendation system. Extensive discussion has been made on the system design, data collection process, data preprocessing, model building, and performance results of the system.

In Chapter 6, *Enhancing Program Management with Predictive Analytics Algorithms (PAAs)*, Bongs Lainjo identifies various practical aspects of software development with a particular focus on predictive model design and deployment. The author also discusses how predictive analytics algorithms can be used to predict future events in different domains of applications, for example, healthcare, manufacturing, education, sports, and agriculture.

I hope that the volume will be very useful for advanced graduate and doctoral students, researchers, practicing data scientists and data engineers, professionals, and consultants working on the broad areas of machine learning, deep learning, artificial intelligence, and other allied areas in computer science. It will also be of use to faculty members and scientists in graduate schools, universities, and research labs engaged in teaching research in machine learning and its applications. Since the volume is not an introductory treatise on the subject, the readers are expected to have basic knowledge of the fundamental principles and theoretical backgrounds of machine learning models and algorithms to understand their advanced applications discussed in the book.

I express my sincere thanks to all authors of the chapters in the volume for their valuable contributions. Without their cooperation and eagerness to contribute, this project would never have been successful. All the authors have been extremely cooperative and punctual during the submission, editing, and publication process of the book. I express our heartfelt thanks to Ms. Marica Novakovic, author service manager of IntechOpen Publishers, for her support, encouragement, patience, and cooperation during the project and her wonderful coordination of all activities involved in it. My sincere thanks also go to Ms. Anja Filipovic, commissioning editor of IntechOpen Publishers, for reposing faith in me and delegating me with the critical responsibility of editorship of such a prestigious academic volume. I will be certainly failing in my duty if I do not acknowledge the encouragement, motivation, and assistance I received from Ms. Sidra Mehtab, my assistant editor and the co-author of one of the chapters in the book. I sincerely thank the valuable contributions received from the coauthors, of the introductory chapter of the book, Mr. Rajdeep Sen and Mr. Abhishek Dutta. The members of my family have always been the sources of my inspiration and motivation for such scholastic work. I dedicate this volume to my beloved sister, Ms. Nabanita Sen, who unfortunately left us on 27 September 2021, due to the deadly disease of cancer. My sister has been always the pillar of strength for me, and this volume also would not have been a success without her constant support and motivation, despite her suffering due to her terminal illness. Last but not the least, I gratefully acknowledge the support and motivation I received from my wife Ms. Nalanda Sen, my daughter Ms. Ritabrata Sen, my mother Ms. Krishna Sen, my brother Mr. Rajdeep Sen, and my sister-in-law, Ms. Toby Kar. Without their support, motivation, and inspiration, the publication of this volume would not have been possible.

**Jaydip Sen**

Professor, Department of Data Science, Praxis Business School, Kolkata, India
