Fog Computing, Deep Learning and Big Data Analytics-Research Directions

ISBN-13: 9789811332081 / Angielski / Twarda / 2019 / 71 str.

C. S. R. Prabhu

Spis treści

1

Introduction

1.1.

A new economy based on IOT emerging by 2015

1.1.1

Emergence of IOT

1.1.2

Smart Cities and IOT

1.1.3

Stages of IOT and Stakeholders

1.1.3.1

Stages of IOT

1.1.3.2

Stakeholders

1.1.3.3

Practical Down Scaling

1.1.4

Analytics

1.1.5

Analytics from the Edge to Cloud [179]

1.1.6

Security and Privacy Issues and Challenges in Internet of Things (IOT)

1.1.7

Access

1.1.8

Cost Reduction

1.1.9

Opportunities and Business Model

1.1.10

Content and Semantics

1.1.11

Data based Business models coming out of IOT

1.1.12

Future of IOT

1.1.12.1

Technology Drivers

1.1.12.2

Future possibilities

1.1.12.3

Challenges and Concerns

1.1.13

Big Data Analytics and IOT

1.1.13.1

Infrastructure for integration of Big Date with IOT

1.2

The Technological challenges of an IOT driven Economy

1.3

Fog Computing Paradigm as a solution

1.4

Definitions of Fog Computing

1.5

Characteristics of Fog computing

1.6

Architectures of Fog computing

1.6.1

Cloudlet Architecture

1.6.2

IoX Architecture

1.6.3

Local Grid’s Fog Computing platform

1.6.4

Parstream

1.6.5

Para Drop

1.6.6

Prismatic Vortex

1.7

Designing a robust Fog computing platform

1.8

Present challenges in designing Fog Computing Platform

1.9

Platform and Applications

1.9.1

Components of Fog Computing Platform

1.9.2

Applications and case studies

1.9.2.1

Health data management and Health care

1.9.2.2

Smart village health care

1.9.2.3

Smart home

1.9.2.4

Smart vehicle and vehicular fog computing

1.9.2.5

Augmented Reality applications

2.

Fog Application management

2.1

Introduction

2.2

Application Management Approaches

2.3

Performance

2.4

Latency Aware Application Management

2.5

Distributed Application Development in Fog

2.6

Distributed Data flow approach

2.7

Resource Coordination Approaches

3

Fog Analytics

3.1

Introduction

3.2

Fog Computing

3.3

Stream data processing

3.4

Stream Data Analytics and Fog computing

3.4.1

Machine Learning for Big Data Stream data and Fog Analytics

3.4.1.1

Supervised Learning

3.4.1.2

Distributed Decision Trees

3.5.1.3

Clustering Methods for Big Data

3.4.1.4

Distributed Parallel Association Rule Mining Techniques for Big Data Scenario

3.4.1.5

Dynamic Association Mining

3.4.2

Deep Learning Techniques

3.4.3

Applications of Deep Learning in Big Data Analytics

3.4.3.1

Semantic Indexing

3.4.3.2

Discriminative Tasks and Semantic Tagging

3.4.4.

Deep Learning Challenges in Big Data Analytics

3.4.4.1

Incremental Learning for Non-Stationary Data

3.4.4.2

High-Dimensional Data

3.4.4.3

Large-Scale Models

3.5

Different Approaches of Fog Analytics

3.6

Comparision

3.7

Cloud Solutions for the Edge Analytics

4

Fog Security and Privary

4.1

Introduction

4.2

Secure Communications in Fog Computing

4.3

Authentication

4.4

Privacy Issues

4.5

User Behaviour Profiling

4.6

Dynamic Fog Nodes and EUs

4.7

Malicious Attacks

4.8

Malicious Insider in the Cloud

4.9

Man in the Middle Attack

4.10

Secured Multi-Tenancy

4.11

Backup and Recovery

5

Research Directions

6

CONCLUSION

References

O autorze

Dr. Chivukula Sree Rama Prabhu has held prestigious positions with Government of India and various Institutions. He retired as the Director General of National Informatics Centre (NIC) Ministry of Electronics and Information Technology Government of India, New Delhi, and has worked in various capacities at Tata Consultancy Services (TCS), CMC, TES and TELCO (now Tata Motors). He was also an international resource faculty for the Programs of APO (Asian Productivity Organization), and represented India on the International Panel at Venture 2004 held by APO at Osaka, Japan. He taught and researched at the University of Central Florida, Orlando and also had a brief stint as a Consultant to NASA Cape Canaveral.

 

Mr. Prabhu was unanimously elected and served as the Chairman of Computer Society of India (CSI), Hyderabad Chapter. He is presently working as an Advisor at KL University, Vijayawada, Andhra Pradesh and as a Director, Research and Innovation at Keshav Memorial Institute of Technology (KMIT), Hyderabad.

 

He obtained his master’s degree in Electrical Engineering with specialization in Computer Science from the Indian Institute of Technology, Bombay after a bachelor’s degree in Electronics and Communication Engineering from Jawaharlal Nehru Technological University, Hyderabad in 1976. He has guided a large number of student research projects at master’s level and has published several papers.

Blurb

This book provides a comprehensive picture of fog computing technology, including of fog architectures, latency aware application management issues with real time requirements, security and privacy issues and fog analytics, in wide ranging application scenarios such as M2M device communication, smart homes, smart vehicles, augmented reality and transportation management.

 

This book explores the research issues involved in the application of traditional shallow machine learning and deep learning techniques to big data analytics. It surveys global research advances in extending the conventional unsupervised or clustering algorithms, extending supervised and semi-supervised algorithms and association rule mining algorithms to big data Scenarios. Further it discusses the deep learning applications of big data analytics to fields of computer vision and speech processing, and describes applications such as semantic indexing and data tagging. Lastly it identifies 25 unsolved research problems and research directions in fog computing, as well as in the context of applying deep learning techniques to big data analytics, such as dimensionality reduction in high-dimensional data and improved formulation of data abstractions along with possible directions for their solutions.