McCalla G., and Greer, J., "Granularity-Based Reasoning and Belief Revision in Student Models", Student Modeling: the Key to Individualized Knowledge-Based Instruction, Spring-Verlag, pp. 39-62 (1994)  Home/Search   Document Details and Download   Summary   Related Articles   Check

....students learn how to account for behavioural and environmental factors seen in the real world. This viewpoint enables creation of larger tutoring systems where a student can drop to a Basic ITT for more detailed interaction but otherwise operate at a relatively coarser grain size. 5. CONCLUSION McCalla and Greer (1991) noted that students seem to reason at many grain sizes and appear to have both deep and shallow knowledge at the same time, especially in case of problem solving abilities. They observed that the relationship of partial knowledge to more complete knowledge is also a granularity relationship and ....

McCalla G.I. & Greer J.E. (1991) Granularity-Based Reasoning and Belief Revision in Student Models. Student Modelling:The Key to Individualized Knowledge-Based Instruction (Eds. Greer J.E & McCalla G.I.), Springer-Verlag, pp 39-62.

....depending upon their internal mental models [Sime, 1993] Granularity hierarchies provide the structure necessary to capture all of these curriculum requirements. 22 3.1. 1 Granularity Hierarchies Granularity refers to the level of detail or perspective from which a concept or entity is viewed [McCalla and Greer, 1994, Coulman et al. 1994] A granularity hierarchy captures different levels of detail in a type of semantic network. This semantic network identifies both abstraction and aggregation relationships in two orthogonal dimensions [McCalla et al. 1992] In order to assume that these two dimensions are ....

McCalla, G. I. and Greer, J. E. (1994). Granularity-- based reasoning and belief revision in student models. In Greer, J. E. and McCalla, G. I., editors, Student Modelling: The Key to Individualized Knowledge--Based Instruction, pages 39--62. Springer--Verlag, Berlin.

....to allow testing of multiple learner traits in one model [2, 7, 6, 3, 1] Each of these papers introduces a novel approach towards testing multiple learner traits. The strengths of these separate notions need to be unified in a single adaptive testing environment. Granularity hierarchies [4] provide the required structure for such a union. Prerequisite relationships can provide test item ordering criteria. Aggregation relationships can be used to break multiple trait components into more fundamental components. Aggregation relationships can also capture multiple model representation ....

....learners different types of questions depending upon their internal mental models [6] Granularity hierarchies provide the structure necessary to capture all of these curriculum requirements. Granularity refers to the level of detail or perspective from which a concept or entity is viewed [4]. A granularity hierarchy captures different levels of detail in a type of semantic network. This semantic network identifies both aggregation and prerequisite relationships in two orthogonal dimensions. The aggregation dimension allows higher level concepts to be broken down into subcomponents. ....

G. I. McCalla and J. E. Greer. Granularity--based reasoning and belief revision in student models. In J. E. Greer and G. I. McCalla, editors, Student Modelling: The Key to Individualized Knowledge--Based Instruction, pages 39--62. Springer--Verlag, Berlin, 1994.

.... student s solution is understood by tracking actions step by step; versus completed solution diagnosis, where a student s solution is understood by comparing it against a set of model solutions, as is done in bug catalogues [Johnson and Soloway, 1985; Johnson, 1990] and granularity based diagnosis [McCalla and Greer, 1994]. Ideally, what is needed is an approach to diagnosis that falls somewhere between these two extremes, that is, an approach that provides the flexibility of performing diagnosis at any point during a student s solution, i.e. intra task plan recognition. There are two main differences between ....

....behaviour, and with incomplete knowledge of plans in a domain. Rather than tuning an existing model of plan recognition for ITS, it might prove feasible to extend an ITS diagnosis system to explicitly address plan recognition. We have taken preliminary steps to recast granularity based recognition [McCalla and Greer, 1994] into a plan recognition system suitable for ITS applications [Koehn, 1994] Granularitybased diagnosis is a method for diagnosis of students strategic misconceptions based on semantic parsing of student behaviour in terms of domain specific strategic concepts, followed by pattern matching of the ....

Gordon I. McCalla and Jim E. Greer. Granularity-Based Reasoning and Belief Revision in Student Models. In J. Greer and G. McCalla (Eds.), Student Models: The Key to Individualized KnowledgeBased Instruction. Springer Verlag, New York, 1994, pages 39-62.

....engineered and implemented as a granularity hierarchy, and discusses the pitfalls encountered in developing a granularity hierarchy for a new domain. 2 Background 2.1 Granularity Hierarchies Granularity refers to the level of detail or perspective from which a concept or entity is viewed. [7, 5] A granularity hierarchy captures different levels of detail in a type of semantic network. This semantic network identifies both abstraction and aggregation relationships in two orthogonal dimensions. 6] Given a set of incomplete information, an entity can be recognized by the granularity ....

McCalla, G. I., and Greer, J. E. Granularity--based reasoning and belief revision in student models. In Student Modelling: The Key to Individualized Knowledge--Based Instruction, J. E. Greer and G. I. McCalla, Eds. Springer--Verlag, Berlin, 1994, pp. 39--62.

....from artificial intelligence. Sometimes these methodologies seem far removed from easy applicability, but, in this domain, the task specificity of the activities carried out in OMS should allow the reasonable adaptation of methodologies like granularity based recognition (as in SCENT, McCalla and Greer, 1994) and model knowledge tracing (as in the LISP Tutor, Corbett and Anderson, 1995) which are explicitly designed for task specific, real world domains. In addition to being multi dimensional, the user models in PHelpS are also multi purpose. Their principal purpose is to help find an appropriate set ....

....of the user models. Since they model authentic and well defined tasks in a real world environment, the usual ambiguity and complexity bedevilling the application of user modelling techniques in, say, dialogue (Kobsa and Wahlster, 1989) and intelligent tutoring (Self, 1990, and Greer and McCalla, 1994) do not appear. Such task orientation is often crucial in achieving success, even in these domains (Grosz and Sidner, 1986, and Vassileva, 1996) It is key here, as well. The third source of power is that the user models, in particular knowledge profiles, are represented at multiple levels of ....

[Article contains additional citation context not shown here]

McCalla, G. I., and Greer, J. E. (1994). Granularity-based reasoning and belief revision in student models.

....underway on incorporating this plan recognition system into HYPHYS, an environment for discovery learning with particle scattering experiments. Introduction An Intelligent Tutoring System (ITS) should provide both an environment for discovery learning and some form of individualized instruction (McCalla Greer, 1994). Thus, it is important for an ITS to recognize that students are trying to accomplish specific goals and, presumably, are designing and carrying out plans for achieving those goals. One artificial intelligence technique to infer user goals and plans from behaviour is plan recognition (cf Andr et ....

....as he or she solves a problem. Before explaining the extensions to granularity, a sample domain for investigating instructional plan recognition will be introduced, and a description of the granularity formalism will be provided. For a more detailed account of granularity based diagnosis, refer to McCalla and Greer (1994). HYPHYS A new domain has been selected to further investigate and evaluate this granularity based plan recognition approach. HYPHYS (HYpothesizing in PHYSics) is a series of increasingly complex microworlds in which Physics scattering experiments can be conducted (Greer, Koehn Rodriguez, ....

[Article contains additional citation context not shown here]

McCalla, G.I. & Greer, J.E. (1994). Granularity-Based Reasoning and Belief Revision in Student Models. In J.

No context found.

McCalla G., and Greer, J., "Granularity-Based Reasoning and Belief Revision in Student Models", Student Modeling: the Key to Individualized Knowledge-Based Instruction, Spring-Verlag, pp. 39-62 (1994)

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC