Real-Time Cloud Computing and Machine Learning Applications
With the emergence of revolutionary technological standards such as 5G and Industry 4.0, real-time applications requiring cloud computing and machine learning are becoming increasingly common. Such applications include real-time scheduling and resource allocation in cloud radio access networks, real-time process monitoring and control in the industrial Internet of Things, network traffic analysis, short-term weather forecasting, and robotics. Given the increase in such applications, several cloud service providers such as Microsoft Azure Machine Learning, IBM Watson, and Google AI have started incorporating Artificial Intelligence (AI) applications on their platforms and providing Analytics as a Service. While it is now simple for users to deploy AI or machine learning algorithms using these cloud platforms, researchers from academia and industry can also develop their own machine learning applications and run them on these platforms to benefit from high processing power and global deployability. The main purpose is to provide in-depth coverage of the programming methodologies and configurations required in developing real-time applications that require machine learning algorithms to be hosted on cloud computing platforms to leverage storage and computing resources.
The real-time applications developed target network traffic analysis and weather forecasting systems. Several machine learning algorithms, namely multiple linear regression, K-Nearest-Neighbours, Multi-Layer-Perceptron, and Convolutional Neural Networks, have been employed in the analysis. The programming languages used include Java, Javascript, HTML5, and MATLAB. Moreover, the Netbeans, Eclipse, and Android Studio IDEs have been used to develop desktop, web, and mobile apps and servlets. The use of several application Programming Interfaces (APIs) to develop the desktop, mobile, and web apps has been fully elaborated. The main cloud platform used for network analysis and weather forecasting systems is the IBM Cloud. Still, Google Firebase and Node.js have also been used in other examples of machine learning applications described in the book. In addition to hosting and running applications on the cloud, the setting up of local servers that can act as fog devices, using client-server sockets and network programming methodologies, has also been explained in detail.
With detailed explanations on all fundamental concepts, programming techniques, and configuration steps in developing cloud-hosted machine learning applications, this book will provide excellent guidance and a full hands-on experience to researchers, professionals, and students working in this field.
Reviews
“The increasing adoption of cloud computing and machine learning for real-time applications makes this textbook very timely and valuable. Its contents are presented in an accessible form, enhanced by many clear and well-illustrated examples. Opening with comprehensive coverage of the fundamentals of both topics, it explains how to integrate them into a real-time network analytics system. Real-time examples enable the reader to gain a clear understanding of the technology and its applications. It is highly recommended for practitioners and researchers alike.” – Peter J. Fleming, Emeritus Professor, Department of Automatic Control and Systems Engineering, The University of Sheffield, UK.
Reviews
There are no reviews yet.