Rickel, J., & Porter, B.W. (1997). Automated modeling of complex systems to answer prediction questions. Artificial Intelligence, 93, pp. 201--260.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....in order to disambiguate sentences (e.g. the agent in an eating event is an animate object, the patient is a food) Answering such questions requires manipulating, rather than applying, the KB rules, relations, and constraints. building qualitative models, for example as performed by Tripel [12]) Tripel needed to introspect to find qualitative rules, rather than do inference with them, to construct is models. 9. Support for Knowledge Entry: Ideally, the KB would not be just a passive database, but would react as the knowledge engineer enters data, for example, it might: identify ....

J. W. Rickel. Automated Modeling of Complex Systems to Answer Prediction Questions. PhD thesis, Dept CS, Univ. Texas at Austin, 1995. (Also available as Tech Rept AI-95-234). http://www.cs.utexas.edu/users/mfkb/papers/rickel-phd.ps.Z. 10

....1 Introduction Modeling and simulation are ubiquitous in scientific and engineering practice. Building a model remains however a difficult and time consuming activity. During the last few years, researchers in the field of automated modeling [Falkenheiner and Forbus, 1991] Levy et al. 1997] [Rickel and Porter, 1997], Nayak, 1995] have addressed the problem of easing the sharing and reuse of models by automating the process of model construction (see [Xia and Smith, 1996] for an overview) Compositional modeling [Falkenheiner and Forbus, 1991] is one of the most effective approach for automated modeling. ....

J. Rickel, and B. Porter. Automated modeling of complex systems to answer prediction questions. Artificial Intelligence, 93, pages 201-260, 1997

....by (Uschold 1991) We believe that metadata (related to an ecological dataset at hand) plays an important role in the way modelers reason when modeling ecological systems, and that we can automatically populate the search space with modeling elements which can be justified in the metadata. (Rickel Porter 1997) reports on automated modeling of complex systems, having plant physiology as the evaluation domain. The system built, called TRIPEL, incorporates a new compositional modeling algorithm which, given a prediction (what if) question, the variables of the physical system, the influences among them, ....

Rickel, J., and Porter, B. 1997. Automated modeling of complex systems to answer prediction questions. Artificial Intelligence Journal 93(1-2):201--260.

....represented, aiming at developing computer tools able to automatically follow modelling principles [Xia and Smith, 1996] Compositional Modelling, the dominant automated modelling approach, consists of assembling basic model fragments, usually gathered in a library, into an welldefined model. In [Rickel and Porter, 1997] we find an application of compositional modelling to an ecology related domain, namely plant physiology, chosen to demonstrate automated modelling of complex systems. The domain knowledge used by TRIPEL, the automated modelling system, is part of a large multipurpose biology knowledge base. The ....

Rickel, J. and Porter, B. (1997). Automated modeling of complex systems to answer prediction questions. Artificial Intelligence Journal 93(1-2):201-260.

....modeler to identify a partitioning of the variables. Providing such a partitioning can simply be viewed as part of the model building task. Automating this process, however, will simplify the model building process and be beneficial when used in conjunction with automated model building systems (Rickel Porter, 1997). The problem of partitioning a graph into closely related components has already been studied extensively in fields such as graph theory (Even, 1979) and constraint satisfaction (Tsang, 1993) Developing a partitioning algorithm for DecSIM primarily requires a characterization of the task so that ....

....and forth. Figure 4.2 demonstrates how chatter affects the simulation results for a model 1 . INF 0 T0 T0 T1 WATER.AMOUNT . INF 0 T0 T0 T1 CROSS SECTIONAL AREA (a) Behavior tree. b) Behavior plots. Rickel and Porter (Rickel Porter, 1997) use qualitative simulation to answer prediction questions within the domain of plant physiology. The TRIPEL algorithm automatically generates a qualitative model from a large scale botany knowledge base (Porter, Lester, Murray, Pittman, Souther, Acker, Jones, 1988) in response to a user query. ....

[Article contains additional citation context not shown here]

Rickel, J., & Porter, B. (1997). Automated modeling of complex systems to answer prediction questions. Artificial Intelligence, 93, 201--260.

....factors a representation into its independent features. Components provide the seams by which a representation can be appropriately disassembled 6 . This ability to generate and work with more abstract domain models is often desirable, both for efficiency and explainability to a user [5, 10]. Work on compositional modeling achieves this advantage in a similar way (e.g. 5, 8, 10] In compositional modeling, a component (called a model fragment ) contributes a set of constraints 7 to a representation of the domain, and composition involves collecting constraints from a suitably ....

.... seams by which a representation can be appropriately disassembled 6 . This ability to generate and work with more abstract domain models is often desirable, both for efficiency and explainability to a user [5, 10] Work on compositional modeling achieves this advantage in a similar way (e.g. [5, 8, 10]) In compositional modeling, a component (called a model fragment ) contributes a set of constraints 7 to a representation of the domain, and composition involves collecting constraints from a suitably chosen set of model fragments. By controlling which model fragments are included in a ....

J. W. Rickel. Automated Modeling of Complex Systems to Answer Prediction Questions. PhD thesis, Dept CS, Univ. Texas at Austin, 1995. (Also available as Tech Rept AI-95-234).

....of dynamically assembling representations is important inspiration for our work. Compositional Modeling A fourth approach to assembling representations comes from work on compositional modeling, the task of building a model of a physical system adequate for answering questions about the system [10, 13, 17]. In compositional modeling, a component (called a model fragment) contains a set of constraints 2 with well defined semantics, rather than being represented in syntactic terms as a partial data structure. For example, a component may represent a resistor by a set of constraints relating the ....

J. W. Rickel. Automated Modeling of Complex Systems to Answer Prediction Questions. PhD thesis, Dept CS, Univ. Texas at Austin, 1995. (Also available as Tech Rept AI-95-234).

....context of ongoing research on knowledge base construction by composition. Elsewhere we have discussed: motivations for the approach and algorithms [5, 6, 7] a graphical user interface [8] a knowledge representation and reasoning system [6] question answering and explanation generation [17, 24] Within that context, this paper provides a brief tour of an early version of our component library to highlight its requirements, construction, contents and applications. In the following section, we will describe our research project in more detail and the design constraints it places on our ....

Rickel, J. and Porter, B. Automated Modeling of Complex Systems to Answer Prediction Questions. Artificial Intelligence Journal 93, 1-2 (1997), 201-260.

....lab. Our language is based on the influences among state variables in a simulation model. Such a representation has been used by human modelers in a wide variety of domains [4, 12, 15] and it has also received considerable attention in research on automated reasoning about physical systems [2, 14]. The details of our language are drawn from Qualitative Process Theory (QPT) 2] which provides a qualitative abstraction of differential equations. The current implementation supports QPT s functional influences among state variables, Q and Q , defined as follows: x Q y , y = f(x; ....

Jeff Rickel and Bruce Porter. Automated modeling of complex systems to answer prediction questions. Artificial Intelligence, 93:201--260, 1997.

....of ongoing research on knowledge base construction by composition. Elsewhere we have discussed: motivations for the approach and algorithms [5, 6, 8] a graphical user interface [8] a knowledge representation and reasoning system [6] question answering methods and explanation generation [18, 25] Within that context, this paper provides a brief tour of an early version of our component library to highlight its requirements, contents, and applications. In the following section, we will describe our research project in more detail and the design constraints it places on our component ....

Rickel, J. and Porter, B. Automated Modeling of Complex Systems to Answer Prediction Questions. Artificial Intelligence Journal 93, 1-2 (1997), 201-260.

....expansion, exhaustively generates the system description before model construction begins. In contrast, we have developed a method that interleaves generating the system description with constructing the model, thereby generating only those parts of the system description that are needed [44]. In this paper, we treat the system description as given, but our modeling algorithm is compatible with either approach. In the compositional modeling approach, elements of the system description serve as building blocks for model construction. We adopt the approach to compositional modeling ....

....separate pools as a single aggregate when they equilibrate on a faster time scale. Thus, a time scale of interest allows many important model simplifications. Although the person posing a question may specify a time scale of interest, often a modeler must determine it automatically. Elsewhere [44, 45], we describe an algorithm for choosing an appropriate time scale of interest when none is specified in the question. Whether the time scale of interest is chosen by the modeler or provided by the person posing the question, this paper will treat it as part of the question. 3 Scenario Models ....

[Article contains additional citation context not shown here]

Jeff Rickel. Automated Modeling of Complex Systems to Answer Prediction Questions. PhD thesis, Department of Computer Science, University of Texas at Austin, May 1995. Technical Report AI95-234.

....base was constructed. Earlier research on one largescale project the Botany knowledge base project (Porter et al. 1988) shows that if detailed, declarative representations of concepts are available, then sophisticated question answering performance can be achieved (Lester Porter 1997; Rickel Porter 1997). However, manually constructing such representations is laborious, and this proved to be a major bottleneck in the project; moreover, it is simply not possible to anticipate all the concept representations that may be needed for answering questions. This points to a fundamental requirement for ....

Rickel, J. W., and Porter, B. W. 1997. Automated modeling of complex systems to answer prediction questions. Artificial Intelligence. (to appear).

No context found.

Rickel, J., & Porter, B.W. (1997). Automated modeling of complex systems to answer prediction questions. Artificial Intelligence, 93, pp. 201--260.

No context found.

Rickel, J., and Porter, B. W. 1997. Automated modeling of complex systems to answer prediction questions. Artificial Intelligence 93:201--260.

No context found.

Rickel, J., & Porter, B.W. (1997).Automated modeling of complex systems to answer prediction questions. Artificial Intelligence, 93, pp. 201--260.

No context found.

Rickel, J. (1997). Automated modeling of complex systems to answer prediction question. Artificial Intelligence 93: 201-260.

No context found.

Rickel, J. and Porter, B. Automated modeling of complex systems to answer prediction questions. Artificial Intelligence, 93:201--260, 1997.

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