China
Africa
Explore chapters and articles related to this topic
Systems for Interpretation and Diagnosis
Published in Adrian A. Hopgood, Intelligent Systems for Engineers and Scientists, 2021
Several different forms of knowledge can contribute to a solution. We have paid specific attention to rules, case histories, and physical models. We have also shown that neural networks and conventional programming can play important roles when included as part of a blackboard system. Rules can be used to represent both shallow (heuristic) and deep knowledge. They can also be used for the generation and verification of hypotheses. Case-based reasoning, introduced in Chapter 5, involves comparison of a given scenario with previous examples and their solutions. Model-based reasoning relies on the existence of a model of the physical system, where that model can be used for monitoring, generation of hypotheses, and verification of hypotheses by simulation.
Supervisory Control and the Design of Intelligent User Interfaces
Published in Raja Parasuraman, Mustapha Mouloua, Automation and Human Performance: Theory and Applications, 2018
Bruce G. Coury, Ralph D. Semmel
Model-based reasoning also provides important insights into the development of our Model Manager. Model-based reasoning relies on a working knowledge of system function and structure, and uses that knowledge to assess or predict system behavior. Well suited to modeling and simulation of physical processes, model-based approaches are commonly found in the areas of diagnosis and control (Dvorak & Kuipers, 1991; Sheridan, 1992). The general idea of a model-based approach is to determine (usually in a hierarchical fashion) which part of the structure of a system is responsible for producing the observed behavior. Such an approach does not rely solely on the mapping of symptoms to faults (and a comprehensive set of cases to handle all contingencies), but instead depends on a simulation or a quantitative model of a dynamic system or problem.
Telecommunications
Published in Jay Liebowitz, The Handbook of Applied Expert Systems, 2019
Model-based reasoning is often used in the telecommunications domains where the dynamics or behavior can be modeled. Inference based on a correct model can be deep, direct, and efficient. The representation of such a model can be tables, finite machines, semantic networks, logic formulas, or Hidden Markov models. Typically, given the observed input, model-based systems use a model to predict the expected behavior, compare it with actual behavior, and then proceed to the next step in the model or stop to draw some conclusion. If a model is available, the correctness and completeness of the model determines the quality of model-based reasoning.
Involving blind user/experts in architectural design: conception and use of more-than-visual design artefacts
Published in CoDesign, 2021
Peter-Willem Vermeersch, Ann Heylighen
Another role of sketches in design relates to (re-)interpreting depicted design ideas (Goldschmidt 1991, Schön & Wiggins 1992). Key to reasoning through models is the dynamic aspect of creative manipulations (Osbeck and Nersessian 2006, following Nersessian 1999). Model-based reasoning is based on the duality between fixing and manipulating: models are constructed to define a problem space, but allow drawing inferences from manipulation. Models are manipulated (physically, mentally, mathematically, gesturally) in the interaction between different persons and artefacts (Osbeck and Nersessian 2006). Manipulation can thus be as temporary as gesturing how to change a certain element in the design. Yet since models also serve to define a problem space, being able to fix some manipulations more permanently is helpful. A design move can be made verbally, by gesturing on the models, or by altering the model itself. All these practices help the cyclic process of definition and reinterpretation (Goldschmidt 1991). In our study, the models provided a means to fix spatial configurations for further interpretation (Vignette 8), invited different kinds of gesturing (see above), and even provided a balance between temporary fixation and haptic use (Vignette 2; model 1).