Device Representation and Reasoning with Affective Relations
نویسندگان
چکیده
ly , reasoning w i t h affective relations proceeds by examin ing the ou tpu t of the device to determine the state of key components, and then propagat ing this i n format ion through the affective relations to deduce the state of other components. Act ions are then performed based on the nature of the alarm messages f rom the device and the states of the components. To make this more concrete the rest of this section presents an example of an affective relat ion model. 2.1 D e v i c e R e p r e s e n t a t i o n The key elements of our device model are 1) a set of basic classes (e.g., Device), 2) a set of diagnostically mot ivated affective relations re lat ing device components, 3) a set of relations relat ing other classes in the model , and 4) a set of properties and general relationships specifying the semantics of the relations, i.e., the knowledge that is impl ic i t in any expl ic i t ly specified knowledge. See Figure 1 for a knowledge level [Newell, 1982] description of a port ion of the device model . (See [ A T T , 1995] for a symbol level description of a much larger por t ion of the device model.) Affective relations l ink components to other components in diagnostically useful ways. For example, i r amed ia te -pa r t -o i means that a direct subcomponent/component relationship exists between two devices. P a r t o f reflects direct as well as transi t ively closed subcomponent/component relations, while its inverse subpa r t reflects direct and transit ive component/subcomponent relations. Depends-on means that the correct funct ioning of a device depends on the correct funct ioning of another device; dependent represents the inverse relation between the two devices. Standby means that if one device fails, its standby partner w i l l take over automat ical ly ; it also means that reliable operation is jeopardized if one device fails. Other relations l ink devices to other classes in the model. S e l f a l a r m means that a device is signaling an a larm. However, a problem w i th one device may not necessarily cause that device to itself signal an a larm; instead, the problem may affect a second device, causing it to signal an a larm. P a r t a l a r m and dependen t -a l a rm l ink devices to alarms signaled by other devices. P a r t a l a r m relates a device to an a larm on another device when the a larming device is a s u b p a r t of the device; dependen t -a l a rm relates devices to alarms based on depends-on. Devices are either f u n c t i o n a l or not. If a device depends on another device that is not funct ional , the dependent device is also not funct ional . If a device is in s i m p l e x mode, it means that none of the device's standbys are funct ional . To i l lustrate the use of these relations, we examine in some detai l a d ig i ta l to l l switching system known as a 4ESS switch. A por t ion of the basic class hierarchy is shown in Figure 2. The class Device is broken down into a series of 4ESS-specific subclasses. Each major type of component of the 4ESS (down to the level of field-replaceable units) adds a class to the general device ontology shown in Figure 1. Figure 3 shows how some of the key port ions of the model of the 4ESS switch are instant iated using the general and domain-specific port ions of the ontology shown We use typewriter font for r e l a t i o n s and sans serif for classes (and their members). To eliminate jargon, hardware component names have been replaced with ABC, etc.
منابع مشابه
A novel model of clinical reasoning: Cognitive zipper model
Introduction: Clinical reasoning is a vital aspect of physiciancompetence. It has been the subject of academic research fordecades, and various models of clinical reasoning have beenproposed. The aim of the present study was to develop a theoreticalmodel of clinical reasoning.Methods: To conduct our study, we applied the process of theorysynthesis in accordan...
متن کاملDevice Representation and Reasoning with A ective Relations
Device representation and reasoning with aaec-tive relations occupies a middle ground between classical model-based diagnosis and heuristic expert systems. A device is modeled by specifying a set of diagnostically motivated aaec-tive relations among its components. Reasoning is then performed by a set of inference rules that reason with the model to propagate symptoms through the components. Re...
متن کاملThe mediating role of anger rumination in relations of affective temperaments with symptoms of conduct disorder in adolescents
Introduction: the temperamental traits are considered as an important risk factor for conduct disorder and the mechanisms of this relations is not clear. The aim of this study was investigating the mediating role of anger rumination in relations of affective temperaments with symptoms of conduct disorder in adolescents Methods: The method of this study was descriptive- correlational. The stati...
متن کاملCritical Analysis of Women’s Representation in TV Advertisements from a Cultural Studies Perspective
Abstract Through the interpretation of texts, and subsequent creation of social reality, mediated representations are often seen to be presented within the certain of ideological discourses that reflect the existing power structures. The main objective of this paper is to analyze television commercials with an emphasis on gender roles to decode the main elements of a dominant discourse (prefer...
متن کاملPredicting Student Performance in Verbal Math Problems Based on Cognitive, Metacognitive, and Affective Factors
Predicting Student Performance in Verbal Math Problems Based on Cognitive, Metacognitive, and Affective Factors F. Karimi, Ph.D. A.R. Moraadi, Ph.D. P. Kadivar, Ph.D. R. Kormi Noori, Ph.D. To determine the predictive role of metacognitive, cognitive, and affective factors in solving verbal math problems, a cluster sample of 450 junior high school students was given ...
متن کامل