Quantitative Semantic Analysis and Comprehension by Cognitive Machine Learning

1. Introduction

It is recognized that a fundamental challenge to machine learning is knowledge learning mimicking the brain (Wang, 2015a, 2016) beyond traditional object identification, cluster classification, pattern recognition, functional regression, and behavior acquisition (Russell & Norvig, 2010; Mehryar et al., 2012; Russell & Norvig, 2010; Wang, 2015a). This leads to an emerging field of cognitive machine learning (Wang, 2010, 2015a, 2016a) on the basis of recent breakthroughs in denotational mathematics (Wang, 2002, 2012, 2013, 2015b) and mathematical engineering (Wang, 2015c, 2016a) such as concept algebra (Wang, 2015b), semantic algebra (Wang, 2013; Wang & Berwick, 2013), inference algebra (Wang, 2011), and Real-Time Process algebra (RTPA) (Wang, 2002). Cognitive machine learning is underpinned by basic studies on cognitive informatics (Wang, 2003, 2007) and cognitive computing (Wang, 2009) such as the Layered Reference Model of the Brain (LRMB) (Wang et al., 2006), the Theory of Abstract Intelligence (αI) (Wang, 2007a), Mathematical Models of Human Memories (Wang, 2007b), Mathematical Models of Cognitive Computing (MMCC) (Wang, 2009), Mathematical Models of Cognitive Neuroinformatics (MMCN) (Wang, 2003), Mathematical Models of Cognitive Linguistics (MMCL) (Wang & Berwick, 2012, 2013), and the Cognitive Models of Brain Informatics (CMBI) (Wang, 2015a; Wang & Wang, 2006).