Page 10 - Data Science Algorithms in a Week

PREFACE

After decades of basic research and more promises than impressive applications,
artificial intelligence (AI) is starting to deliver benefits. A convergence of advances is
motivating this new surge of AI development and applications. Computer capability as
evolved from high throughput and high performance computing systems is increasing. AI
models and operations research adaptations are becoming more matured, and the world is
breeding big data not only from the web and social media but also from the Internet of
Things.
This is a very distinctive book which discusses important applications using a variety
of paradigms from AI and outlines some of the research to be performed. The work
supersedes similar books that do not cover as diversified a set of sophisticated
applications. The authors present a comprehensive and articulated view of recent
developments, identifies the applications gap by quoting from the experience of experts,
and details suggested research areas.
The book is organized into 14 chapters which provide a perspective of the field of AI.
Areas covered in these selected papers include a broad range of applications, such as
manufacturing, autonomous systems, healthcare, medicine, advanced materials, parallel
distributed computing, and electronic commerce. AI paradigms utilized in this book
include unsupervised learning, ensembles, neural networks, deep learning, fuzzy logic,
support-vector machines, genetic algorithms, genetic programming, particle swarm
optimization, agents, and case-based reasoning. A synopsis of the chapters follow:
• Clustering Techniques: Novel research in clustering techniques are essential to
improve the required exploratory analysis for revealing hidden patterns, where label
information is unknown. Ramazan Ünlü in the chapter “Unsupervised Ensemble
Learning” discusses unsupervised ensemble learning, or consensus clustering which is a
method to improve the selection of the most suitable clusterization algorithm. The goal of
this combination process is to increase the average quality of individual clustering
methods. Through this chapter, the main concepts of clustering methods are introduced