Big-Data Analytics for Cloud, IoT and Cognitive Computing

Kai Hwang, Min Chen

• Publisher: Wiley
• Year: 2017
• Print ISBN: 9781119247029
• eBook ISBN: 9781119247296
• Edition: 1
• Topic: Computer Science
• Subtopic: Computer Science General
• Index: Computer Science

The definitive guide to successfully integrating social, mobile, Big-Data analytics, cloud and IoT principles and technologies

The main goal of this book is to spur the development of effective big-data computing operations on smart clouds that are fully supported by IoT sensing, machine learning and analytics systems. To that end, the authors draw upon their original research and proven track record in the field to describe a practical approach integrating big-data theories, cloud design principles, Internet of Things (IoT) sensing, machine learning, data analytics and Hadoop and Spark programming.

Part 1 focuses on data science, the roles of clouds and IoT devices and frameworks for big-data computing. Big data analytics and cognitive machine learning, as well as cloud architecture, IoT and cognitive systems are explored, and mobile cloud-IoT-interaction frameworks are illustrated with concrete system design examples. Part 2 is devoted to the principles of and algorithms for machine learning, data analytics and deep learning in big data applications. Part 3 concentrates on cloud programming software libraries from MapReduce to Hadoop, Spark and TensorFlow and describes business, educational, healthcare and social media applications for those tools.

• The first book describing a practical approach to integrating social, mobile, analytics, cloud and IoT (SMACT) principles and technologies
• Covers theory and computing techniques and technologies, making it suitable for use in both computer science and electrical engineering programs
• Offers an extremely well-informed vision of future intelligent and cognitive computing environments integrating SMACT technologies
• Fully illustrated throughout with examples, figures and approximately 150 problems to support and reinforce learning
• Features a companion website with an instructor manual and PowerPoint slides www.wiley.com/go/hwangIOT

Big-Data Analytics for Cloud, IoT and Cognitive Computing satisfies the demand among university faculty and students for cutting-edge information on emerging intelligent and cognitive computing systems and technologies. Professionals working in data science, cloud computing and IoT applications will also find this book to be an extremely useful working resource. 

Part 1
Big Data, Clouds and Internet of Things

1
Big Data Science and Machine Intelligence

---

CHAPTER OUTLINE

1. 1.1 Enabling Technologies for Big Data Computing
   1. 1.1.1 Data Science and Related Disciplines
   2. 1.1.2 Emerging Technologies in the Next Decade
   3. 1.1.3 Interactive SMACT Technologies
2. 1.2 Social-Media, Mobile Networks and Cloud Computing
   1. 1.2.1 Social Networks and Web Service Sites
   2. 1.2.2 Mobile Cellular Core Networks
   3. 1.2.3 Mobile Devices and Internet Edge Networks
   4. 1.2.4 Mobile Cloud Computing Infrastructure
3. 1.3 Big Data Acquisition and Analytics Evolution
   1. 1.3.1 Big Data Value Chain Extracted from Massive Data
   2. 1.3.2 Data Quality Control, Representation and Database Models
   3. 1.3.3 Big Data Acquisition and Preprocessing
   4. 1.3.4 Evolving Data Analytics over the Clouds
4. 1.4 Machine Intelligence and Big Data Applications
   1. 1.4.1 Data Mining and Machine Learning
   2. 1.4.2 Big Data Applications – An Overview
   3. 1.4.3 Cognitive Computing – An Introduction
5. 1.5 Conclusions

---

1.1 Enabling Technologies for Big Data Computing

Over the past three decades, the state of high technology has gone through major changes in computing and communication platforms. In particular, we benefit greatly from the upgraded performance of the Internet and World Wide Web (WWW). We examine here the evolutional changes in platform architecture, deployed infrastructures, network connectivity and application variations. Instead of using desktop or personal computers to solve computational problems, the clouds appear as cost-efficient platforms to perform large-scale database search, storage and computing over the Internet.

This chapter introduces the basic concepts of data science and its enabling technologies. The ultimate goal is to blend together the sensor networks, RFID (radio frequency identification) tagging, GPS services, social networks, smart phones, tablets, clouds and Mashups, WiFi, Bluetooth, wireless Internet+, and 4G/5G core networks with the emerging Internet of Things (IoT) to build a productive big data industry in the years to come. In particular, we will examine the idea of technology fusion among the SMACT technologies.

1.1.1 Data Science and Related Disciplines

The concept of data science has a long history, but only recently became very popular due to the increasing use of clouds and IoT for building a smart world. As illustrated in Figure 1.1, today's big data possesses three important characteristics: data in large volume, demanding high velocity to process them, and many varieties of data types. These are often known as the five V's of big data, because some people add two more V's of big data: one is the veracity, which refers to the difficulty to trace data or predict data. The other is the data value, which can vary drastically if the data are handled differently.

Figure 1.1 Big data characteristics: Five V's and corresponding challenges.

By today's standards, one Terabyte or greater is considered a big data. IDC has predicted that 40 ZB of data will be processed by 2030, meaning each person may have 5.2 TB of data to be processed. The high volume demands large storage capacity and analytical capabilities to handle such massive volumes of data. The high variety implies that data comes in many different formats, which can be very difficult and expensive to manage accurately. The high velocity refers to the inability to process big data in real time to extract meaningful information or knowledge from it. The veracity implies that it is rather difficult to verify data. The value of big data varies with its application domains. All the five V's make it difficult to capture, manage and process big data using the existing hardware/software infrastructure. These 5 V's justify the call for smarter clouds and IoT support.

Forbes, Wikipedia and NIST have provided some historical reviews of this field. To illustrate its evolution to a big data era, we divide the timeline into four stages, as shown in Figure 1.2. In the 1970s, some considered data science equivalent to data logy, as noted by Peter Naur: “The science of dealing with data, once they have been established, while the relation of the data to what they represent is delegated to other fields and sciences.” At one time, data science was regarded as part of statistics in a wide range of applications. Since the 2000s, the scope of data science has become enlarged. It became a continuation of the field of data mining and predictive analytics, also known as the field of knowledge discovery and data mining (KDD).

Figure 1.2 The evolution of data science up to the big data era.

In this context, programming is viewed as part of data science. Over the past two decades, data has increased on an escalating scale in various fields. The data science evolution enables the extraction of knowledge from massive volumes of data that are structured or unstructured. Unstructured data include emails, videos, photos, social media, and other user-generated contents. The management of big data requires scalability across large amounts of storage, computing and communication resources.

Formally, we define data science as the process of extraction of actionable knowledge directly from data through data discovery, hypothesis and analytical hypothesis. A data scientist is a practitioner who has sufficient knowledge of the overlapping regimes of expertise in business needs, domain knowledge, analytical skills and programming expertise to manage the end-to-end scientific process through each stage in the big data life cycle.

Today's data science requires aggregation and sorting through a great amount of information and writing algorithms t...

Table of contents

• Cover
• Title Page
• Copyright
• About the Authors
• Preface
• About the Companion Website
• Part 1 Big Data, Clouds and Internet of Things
• Part 2 Machine Learning and Deep Learning Algorithms
• Part 3 Big Data Analytics for Health-Care and Cognitive Learning
• Index
• EULA