Textbook in PDF format

This volume provides the reader with a comprehensive and up-to-date treatise positioned at the junction of the areas of Machine Learning (ML) and Granular Computing (GrC). ML offers a wealth of architectures and learning methods. Granular Computing addresses useful aspects of abstraction and knowledge representation that are of importance in the advanced design of ML architectures. In unison, ML and GrC support advances of the fundamental learning paradigm. As built upon synergy, this unified environment focuses on a spectrum of methodological and algorithmic issues, discusses implementations and elaborates on applications. The chapters bring forward recent developments showing ways of designing synergistic and coherently structured ML-GrC environment. The book will be of interest to a broad audience including researchers and practitioners active in the area of ML or GrC and interested in following its timely trends and new pursuits.
Machine Learning has been an intensive research endeavor leading in recent years to a wealth of concepts, algorithms, and implementations encompassing a variety of original and far-reaching application domains. The successes of designed learning environments are highly impactful, especially in the realm of natural language processing (NLP) as well as image processing and computer vision.
Granular Computing realizes fundamental ideas of abstraction: using information granules we describe the essence of experimental data through a collection of meaningful information granules. The granules deliver an abstract representation being essential to reveal and capture the available knowledge at a suitable level of detail, organize knowledge, realize processing, and facilitate communication of findings. Numerous formal frameworks producing the operational realization of information granules in the form of sets (intervals), fuzzy sets, and probabilistic information granules, among others, deliver a sound computing environment with all processing faculties.
In a nutshell, in virtue of their research agendas, granular computing and Machine Learning are poised to build a coherent and highly synergistic environment and establish a series of critically beneficial linkages. Owing to the abilities of represent knowledge, capturing aspects of information granularity, building various layers of abstraction, granular computing offers interesting conceptual and computational vehicle to cope with open and central aspects of machine learning including building front and back-end layers of learning architectures, address the issues of computing overhead, dealing with privacy questions, address interpretability and explainability concerns, resolve quests of credibility of machine learning constructs and ensuing results, and building mechanisms of self-awareness of the developed classifiers, predictors, or associators. The learning paradigms and learning practices could become indispensable in building architectures of granular computing and endowing them with optimization mechanisms helpful in the construction of information granules