China 
Africa 
Explore chapters and articles related to this topic
Employing Ontology to Capture Expert Intelligence within GEOBIA: Automation of the Interpretation Process
Published in Raechel A. White, Arzu Çöltekin, Robert R. Hoffman, Remote Sensing and Cognition, 2018
Sachit Rajbhandari, Jagannath Aryal, Jon Osborn, Arko Lucieer, Robert Musk
Classification in object-based image analysis is carried out using domain knowledge. The formalization of knowledge provides an insightful and systematic approach to accessing the information necessary to identify image objects. An ontology defines concepts and their relationships for knowledge representation and reasoning. In the context of remote sensing images, ontology helps to conceptualize image concepts (e.g., land, forest, water bodies, etc.) based on spectral, spatial, and textural features and link them to expert knowledge. A framework is proposed by Andres et al. (2012), where an image ontology is developed to define the core concepts of the image. Formally defined image concepts are used to describe segmented images with low-level features and to build an ontology that represents high-level concepts in the images to classify image segments. The need for formalizing expert knowledge for the automatic semantic interpretation of remote sensing images is highlighted in the study by Andrés, Pierkot, and Arvor (2013).
Parallel and Distributed Architecture for Intelligent Systems
Published in Konar Amit, Artificial Intelligence and Soft Computing, 2018
Distributed representation of knowledge is preferred for enhancing parallelism in the system. A Petri net, for example, is one of the structural models, where each of the antecedent and the consequent clauses are represented by places and the if-then relationship between the antecedent and consequent clauses are represented by transitions. With such representation, a clause denoted by a place may be shared by a number of rules. Distribution of fragments of a knowledge on physical units (here places and transitions) enhances the degree of fault tolerance of the system. Besides Petri nets, other connectionist approaches for knowledge representation and reasoning include neural nets, frames, semantic nets and many others.
Synthetic Worlds for On-Demand Experience
Published in C.A.P. Smith, Kenneth W. Kisiel, Jeffrey G. Morrison, Working Through Synthetic Worlds, 2009
Knowledge representation and reasoning has been at the core of AI since the beginning and is concerned with how knowledge can be represented symbolically and manipulated in an automated way by reasoning programs (Brachman 2004). Since knowledge is used to achieve intelligent behavior, the fundamental goal of knowledge representation is to represent knowledge in a manner as to facilitate reasoning or inferencing (i.e., drawing conclusions) from knowledge. The key problem has always been to find a representation scheme (and a supporting reasoning system) that can make the needed inferences within time and resource constraints.
The effect of voice and humour on users’ perceptions of personal intelligent agents
Published in Behaviour & Information Technology, 2021
Sara Moussawi, Raquel Benbunan-Fich
This paper is concerned with the investigation of personal intelligent agents, which are also referred to as intelligent personal assistants, smart personal assistants, virtual assistants, digital assistants and conversational agents in the literature. We define PIAs as software that acts intelligently and uses natural language to assist a human (Russell and Norvig 2010; March, Hevner, and Ram 2000; Shoham 1993; Wooldridge and Jennings 1995). PIAs are different from other systems previously explored in IS research and are distinguished by unique characteristics that enhance users’ perceptions of their intelligence (based on natural language processing abilities, knowledge representation, automated reasoning and machine learning capabilities), and anthropomorphism (based on social and emotional capacities rather than simply physical features) (Russell and Norvig 2010; Moussawi and Koufaris 2019). PIAs are designed to act with autonomy, pro-activeness, awareness of the environment, and an ability to communicate with the user using natural language, which leads users to form perceptions of intelligence of these systems. These agents possess anthropomorphic features such as voice and humour causing users to develop perceptions of their human-likeness. Existing research aiming to classify PIAs by design characteristics revealed additional attributes including embodiment through human-like characteristics and anthropomorphism, communication with the environment via sensors and natural language, and collective intelligence rooted in improvements in machine learning (Knote et al. 2019).
Featured risk evaluation of nautical navigational environment using a risk cloud model
Published in Journal of Marine Engineering & Technology, 2020
Yan-Fei Tian, Li-Jia Chen, Li-Wen Huang, Jun-Min Mou
Scholars Li et al. proposed a membership cloud model (or cloud model) (Li et al. 1995), that combined the basic principles of probability theory and fuzzy set theory for application in the field of intelligent control (Li et al. 1998). That cloud model fully integrated fuzziness and randomness and provided a mapping between qualitative and quantitative information, and thereby functioned as a powerful tool for processing a combination of qualitative and quantitative information (Wang et al. 2010). More objectively, methods of knowledge representation and reasoning that are based on a cloud model are able to fully express the fuzziness and randomness of uncertainty and, to a certain extent, cloud model based handling methods can resolve missing information while gathering data (Wang and Liu 2012). Over time, the cloud model has been used in an expanded array of applications, and is now widely applied in the comprehensive evaluation of complex systems (Du et al. 2008).
Special issue on robot vision for dexterous manipulation and interaction
Published in Advanced Robotics, 2019
On a semantic level, the first paper by Vitucci and Gini treats the challenge of deductive reasoning in the field of robot manipulation. They propose specific description languages as logic-based frameworks for knowledge representation and reasoning for the scenarios of grasping ontology and object recognition ontology. An integrated concept for pose estimation of stacked objects for mobile manipulation in practical industrial application scenarios in the context of autonomous packaging, specifically picking up of stacked non-rigid objects, is presented in the second paper by Lim et al. Their work solves Challenge 2 of the European Robotics Challenges (EuRoC) project being applied to a realistic industrial environment. The third paper by Zhang-Xu and Kühnlenz investigates visual attention and motion coordination in teams and transfers knowledge gained from human user studies to robotics by proposing a behavior-based approach to attention with actor and observer switching.