Shrager J and Langley, P (1990) (Eds.) Computational Models of Scientific Discovery and Theory Formation, San Mateo: Morgan Kaufman.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....is accepted to the theory; in misspelled words, the new evidence (or novelty) must be consiliatedby the theory, but never quoted extensionally. In this way, our differentiation between Enumerative (or descriptional) Induction and Best Explanation [Ernis 1968] Harman 1965] Hempel 1965] see [Sharger Langley 1990] for more modem contrasted positions in this debate) is not based on the predictive value of a theory but on the usability (in terms of a coherent explanation of reality) and the applicability of abduction. Moreover, in explanatory induction, deduction can and must play a very important role. The ....

Shrager, J.; Langley, P. "Computational Models of Scientific Discovery and Theory Formation" Morgan Kaufmman, 1990.

....Theoretical Level Observational Level Theory Formation Empirical Generalisation Observation Prediction Figure 1 Theory Development 1 Once existing observations and empirical generalisations have been structured, theory formation starts. Procedures, such as the ones described in [shr90] should be of aid here. Moreover, the conclusion that the software process is a feedback system suggests that theory formation may be informed by control theoretic ideas. The discovery of which theory best explains the empirical observations and what its attributes are will provide clues to the ....

Shrager J and Langley P (eds.), Computational Models of Scientific Discovery and Theory Formation, Morgan Kaufmann Publishers, Inc, San Mateo, CA, 1990, 498 pps.

....representations, formal methods and programming languages does exist and may prove important in supporting the proposed study. But such theory is qualitatively different to that being proposed here. As a phenomenological descriptor the latter is more akin to the theories of the physical sciences [shr90]. The basis for axiomatic theory development as epitomised by Euclidean geometry and used for many centuries in the mathematical and physical sciences is well established. So is the application of formal methods and of the many representations and logics in computer science [tur87] Success in ....

Shrager J and Langley P (eds.), Computational Models of Scientific Discovery and Theory Formation, Morgan Kaufmann Publishers, Inc, San Mateo, CA, 1990, 498 pps.

....of the hypothesis can be obtained in a Bayesian way. Since this initial distribution is generally unknown, many different measures of the quality of theories have been proposed in philosophy of science and Machine Learning (ML) generally in an informal way. From these, there are two main trends ([39]) descriptional induction ( 5] which is usually related to the simplicity criterion (or Occam s Razor) and the view of learning as compression; and explanatory induction ( 39] which is more closely related to coherence, cohesion or consilience criteria ( 41] In 1978, Rissanen formalised ....

....in philosophy of science and Machine Learning (ML) generally in an informal way. From these, there are two main trends ( 39] descriptional induction ( 5] which is usually related to the simplicity criterion (or Occam s Razor) and the view of learning as compression; and explanatory induction ([39]) which is more closely related to coherence, cohesion or consilience criteria ( 41] In 1978, Rissanen formalised Occam s Razor under the Minimum Description Length (MDL) principle, quickly spreading over the theory and practice of ML and predictive modelling. In his later formulation ( 5] ....

[Article contains additional citation context not shown here]

Shrager, J.; Langley, P., Computational Models of Scientific Discovery and Theory Formation, Morgan Kaufmman, 1990.

....any novel discoveries, its results are suprising given the simplicity of the design. Keywords: Artificial agents, experiment design, hypothesis testing, statistical analysis. 1 Introduction Planned, controlled discovery of new knowledge is among the most complex of intelligent activities [5,6]. It requires a broad base of knowledge, sound logic, efficient heuristic reasoning, controlled search, and planning skills. Duplicating scientific behavior in an automated agent is a difficult task, given the extraordinary complexity of the problem, but is possible within the framework of current ....

Jeff Shrager and Pat Langley, editors. Computational Models of Scientific Discovery and Theory Formation. Morgan Kaufmann, 1990.

....between them. Since there remains a suspicion that the phases are not defined properly, such a definition should also be examined empirically. Research on automated scientific discovery has addressed many of these issues, and contributed greatly to our understanding of the scientific process[4,8]. The term scientific method has been applied to this process for centuries, but we have recently recognized that the most critical portions are not methodic at all. Rather, interesting discoveries are usually first encountered by the intuition, and then followed by months or years of research in ....

Jeff Shrager and Pat Langley, editors. Computational Models of Scientific Discovery and Theory Formation. Morgan Kaufmann, 1990.

....difficult to understand compared to decision trees. KDD also emphasizes scaling and robustness properties of modeling algorithms for large noisy data sets. Related AI research fields include machine discovery, which targets the discovery of empirical laws from observation and experimentation (Shrager and Langley 1990) (see Kloesgen and Zytkow [1996] for a glossary of terms common to KDD and machine discovery) and causal modeling for the inference of causal models from data (Spirtes, Glymour, and Scheines 1993) Statistics in particular has much in common with KDD (see Elder and Pregibon [1996] and Glymour et ....

Shrager, J., and Langley, P., eds. 1990. Computational Models of Scientific Discovery and Theory Formation.

....express dissatisfaction with the non reflective and often uninteresting nature of their classes project work in general (anecdotal observation from initial SIBLE classroom work) Radinsky, Spillane final March 4, 1999 3 We frame this family of problems as a lack of reflective inquiry. Using D. Kuhn s frame of reference (Kuhn, 1989), we believe that students can develop the ability to more fully differentiate evidence from theory through inquiry, if they are given repeated opportunities to investigate phenomena of interest, generate explanations, and reflectively consider their activities and explanations throughout the ....

In J. Shrager & P. Langley (Eds.), Computational models of scientific discovery and theory formation (pp. 355-402). Palo Alto, CA: Morgan Kaufmann Publishers, Inc. Kuhn, D. (1989). Children and adults as intuitive scientists. Psychological Review, 96, 674-689.

....to suspend and come back to goals based on experimentation and social interaction. Bell s case study shows that an inventor that acts in society must necessarily be opportunistic within the framework supplied by her or his thematic goals and resulting network of enterprises. Therefore, to use Shrager and Langley s (1990) term, our computational model takes a large step towards dealing with the problem of embeddedness. ALEC s architecture grew up from the difficulties encountered with IMPROVISER s opportunistic control (Wills Kolodner, 1994) Simina and Kolodner (1995) postulated the existence of active ....

Shrager, J., & Langley, P. (Eds.) (1990). ComputationalModels of Scientific Discovery and Theory Formation. Morgan Kaufmann.

....and the effects of background knowledge (see Keane, 1988, 1990) 3.2 Models of Analogical Mapping The three main cognitive models of analogical mapping in the literature all instantiate the above informational constraints to varying degrees and some also include behavioural constraints. Falkenhainer et al. s (1986, 1989) Structure Mapping Engine#(SME) implements both structural and similarity constraints in a serial way. SME finds all the legal local matches between two domains and then combines these into alternative interpretations of the comparison. SME is explicitly designed to construct all possible ....

In Shrager, J. & Langley, P. (Eds.), Computational Models of Scientific Discovery and Theory Formation. San Mateo, CA: Morgan Kaufmann. Falkenhainer, B., Forbus, K.D., & Gentner, D. (1986). Structure-mapping engine. Proceedings of the Annual Conference of the American Association for Artificial Intelligence.

.... Warsaw University Banacha 2, 02 097 Warsaw, Poland e mail: skowron mimuw.edu.pl synak mimuw.edu.pl Abstract The problem of decision rules extracting (or more general knowledge discovery) from experimental data is intensively studied (see e.g. 2] 3] 9] 11] 12] 14] 18] 20] [21], 22] 27] We apply rough set methods and boolean reasoning for decision rules discovery from decisions tables. It is not possible in general to extract general laws from experimental data by computing first all reducts of a decision table representing data and next decision rules from these ....

....Key words: evolutionary computation, knowledge discovery, rough sets, decisions algorithms. 1 Introduction The aim of the paper is to present a method for extracting decision rules from experimental data. This problem is intensively studied (see e.g. 2] 3] 9] 11] 12] 14] 18] 20] [21], 22] 27] Our approach is based on rough set methods. The rough set methods [16] developed so far are not always sufficient for extracting laws from decision tables (used to represent data) One of the reasons is that these methods are not taking into account that part of the reduct set is ....

Shrager, J., Langley, P.: Computational models of scientific discovery and theory formation, Morgan Kaufmann, San Mateo 1990.

....a theorem in one of those sciences would be equivalent to solving a system of linear equations. 7 Related work The main difference between boole2 and other systems dealing with the discovery of historical laws, such as bacon, glauber, stahl and dalton, 24,25] am and eurisko [26 28] kekada [20,33], are [32] or cdp [1, pp.46 57] is that boole2 is exclusively guided by theory, instead of experimentation, i.e. it is not a data driven discovery system. boole2 s input is not experimental data, but an abstract representation of some science. Further, many of its heuristics stem from theory, ....

Jeff Shrager and Pat Langley. Computational Models of Scientific Discovery and Theory Formation. Morgan Kaufmann, 1990.

....operation plus the context provided by all previous operations. The search space grows explosively. While research in statistics and artificial intelligence has addressed issues in the automation of later stages of analysis, such as theory generation, model selection, and experiment design [23], less attention has been given to initial exploration of data. We have developed a novel approach to exploration as search. This paper gives an overview of the design of aide, the Assistant for Intelligent Data Exploration, which assists humans in the early stages of data analysis [1] The system ....

Jeff Shrager and Pat Langley. Computational Models of Scientific Discovery and Theory Formation. Morgan Kaufman, 1990.

....Data analysis plays a central role in our attempts to understand the behavior of complex systems. While research in both statistics and artificial intelligence has addressed issues in the automation of later stages of analysis, such as theory generation, model selection, and experiment design [7], less attention has been given to initial exploration of data. Deriving structure from data is nevertheless a necessary first step. Exploratory data analysis (EDA) 8] provides a wide range of statistical tools for looking at data. Human analysts find it straightforward to select and apply these ....

Jeff Shrager and Pat Langley. Computational Models of Scientific Discovery and Theory Formation. Morgan Kaufman, 1990.

.... Computational scientific discovery is no exception, as early research focused on replicating discoveries from the history of disciplines as diverse as mathematics (Lenat, 1977) physics (Langley, 1981) chemistry ( Zytkow Simon, 1986) and biology (Kulkarni Simon, 1990) As the collection by Shrager and Langley (1990) reveals, these efforts also had considerable breadth in the range of scientific activities they attempted to model, though most work aimed to replicate the historical record only at the most abstract level. Despite the explicit goals of this early research, some critics (e.g. Gillies, 1996) have ....

Shrager, J., & Langley, P. (Eds.) (1990). Computational models of scientific discovery and theory formation. San Francisco: Morgan Kaufmann.

.... Computational scientific discovery is no exception, as early research focused on replicating discoveries from the history of disciplines as diverse as mathematics (Lenat, 1977) physics (Langley, 1981) chemistry ( Zytkow Simon, 1986) and biology (Kulkarni Simon, 1990) As the collection by Shrager and Langley (1990) reveals, these efforts also had considerable breadth in the range of scientific activities they attempted to model, though most work aimed to replicate the historical record only at the most abstract level. Despite the explicit goals of this early research, some critics (e.g. Gillies, 1996) have ....

Shrager, J., & Langley, P. (Eds.) (1990). Computational models of scientific discovery and theory formation. San Francisco: Morgan Kaufmann.

No context found.

Shrager J and Langley, P (1990) (Eds.) Computational Models of Scientific Discovery and Theory Formation, San Mateo: Morgan Kaufman.

No context found.

Shrager, J., & Langley, P. (Eds.) (1990). Computational models of scientific discovery and theory formation.

No context found.

Shrager, J., & Langley, P. (Eds.) (1990). Computational models of scientific discovery and theory formation.

No context found.

J. Shrager, P. Langley (eds.) Computational Models of Scientific Discovery and Theory Formation. San Mateo: Morgan Kaufmann Publishres, Inc., 1990.

No context found.

Shrager, J., Langley, P.(1990). "Computational models of scientific discovery and theory formation", Morgan Kaufmann, San Mateo, CA. 27

No context found.

J. Shrager, P. Langley (1990). Computational models of scientific discovery and theory formation, San Mateo, California: Morgan Kaufmann.

No context found.

Langley, Pat and Shrager, Jeff, ed. Computational Models of Scientific Discovery and Theory Formation. Morgan Kaufmann Publishers; San Mateo, CA, 1990.

No context found.

Shrager & P. Langley (Eds.), Computational models of scientific discovery and theory formation. San Mateo, CA: Morgan Kaufmann. Kippenhahn, R., & Weigert, A. (1994). Stellar structure and evolution. London: Springer-Verlag.

No context found.

Shrager & P. Langley (Eds.), Computational models of scientific discovery and theory formation. San Mateo, CA: Morgan Kaufmann. Langley, P. (1981). Data-driven discovery of physical laws. Cognitive Science, 5 , 31--54.

First 50 documents

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