Deep learning for cognitive computing systems is a bird’s-eye view of deep learning and the neighboring, whether loosely or strongly coupled, areas. It is a collection of several authors’ publications.
The first chapter overviews the fundamental concepts of deep learning and the basic architecture models. There is a comparison with the more traditional artificial neuron networks (ANNs), followed by a brief overlook of the application areas. This chapter is a good introductory text on deep learning for a first lecture in a university course.
The second chapter aims to cover the concept of cognitive computing. However, it does not explain the difference between artificial intelligence (AI) and the subfields of AI that provide the algorithmic background, namely machine learning, deep learning, and data analysis. User involvement and integration are not sufficiently emphasized, and cognitive resonance, cognitive adaptation, and cognitive info-communication are not addressed. Figure 4 contains Spanish text for an explanation of the components of the illustration that should have been translated. The chapter gives an outlook on possible application fields, too.
The third chapter overviews the issues of computer vision in light of AI and deep learning. The chapter talks about the relevant algorithms and the broad application areas, for example, the topic of autonomous cars and FinTech. The specificity of image processing is mentioned, and the same set of algorithms is used for deep learning though different data processing, preparation, and data models are required.
The fourth chapter looks at voice recognition systems. Voice recognition systems demand a different data preparation, and the traditional signal processing methods are used to ingest the data and pass it to the adequate algorithms. The wide range of application areas is briefly covered.
Chapters 5 to 10 analyze the various healthcare-related applications. Chapter 9 investigates the possible solutions for cognitive authentication in healthcare. Chapter 6 deals with the Internet of Things (IoT) and its application in healthcare. The security problems and the related authentication are interrelated to the application of IoT in healthcare, namely remote diagnostics, telemedicine, and so on. Information communication and its security play an important role; however, the authors are ignorant of the research that has already been done or is currently being done in the field of cognitive info-communication [1]. The text is hardly understandable at times, and the message of how electroencephalogram (EEG) signals can be used in a real-life application for the authentication of an individual does not come through.
The last chapter is about text summarization and extraction. This research paper includes a proposed solution to the issue. It describes the model and an investigation of its goodness and correctness according to research standards.
The book and its included chapters yield an overarching view of generic deep learning and specific application areas. The chapters do not contain any complicated details. Each chapter gives a list of the significant algorithms and relevant application areas, only lightly touching on the domains analyzed. The book could work for business analysts, business architects, and managers involved in innovative service design who need a superficial survey of the specific fields without the heavy details. I am not a native English speaker; however, even in this case, it seems as though the text contains some strange word choices and sometimes the sentences are not clear. More thorough proofreading would have been beneficial for the book.
