Quantitative and Qualitative Approaches to Reasoning under Uncertainty in Medical Decision Making
|
|
- Kathleen Sparks
- 5 years ago
- Views:
Transcription
1 Quantitative and Qualitative Approaches to Reasoning under Uncertainty in Medical Decision Making John Fox, David Glasspool, and Jonathan Bury Imperial Cancer Research Fund Labs Lincoln s Inn Fields London WC2A 3PX United Kingdom jf@acl.icnet.uk; dg@acl.icnet.uk; jb@acl.icnet.uk Abstract. Medical decision making frequently requires the effective management and communication of uncertainty and risk. However a tension exists between classical probability theory, which is precise and rigorous but which people find non-intuitive and difficult to use, and qualitative approaches which are ad hoc but can be more versatile and easily comprehensible. In this paper we review a range of approaches to uncertainty management, then describe a logical approach, argumentation, which subsumes qualitative as well as quantitative representations and has a clear formal semantics. The approach is illustrated and evaluated in five decision support applications. 1 Introduction Representing and managing uncertainty is central to understanding and supporting much clinical decision-making and is extensively studied in AI, computer science and psychology. Implementing effective decision models for practical clinical applications presents a dilemma. On the one hand, informal and qualitative representations of uncertainty may be natural for people to understand but they often lack formal rigour. On the other hand formal approaches based on probability theory are precise but can be awkward and non-intuitive to use. While in many cases probability theory is an ideal for optimal decision-making it is often impractical. In this paper we review a range of approaches to uncertainty representation, then describe a logical approach, argumentation, which can subsume both qualitative and quantitative approaches within the same formal framework. We believe that this may enable improvements in both the scope and comprehensibility of decision support systems and illustrate the approach with five decision support applications developed by our group. 2 Decision Theory and Decision Support It is generally accepted that human reasoning and decision-making can exhibit various shortcomings when compared with accepted prescriptive theories derived from mathematical logic and statistical decision-making, and there are systematic patterns of distortion and error in people s use of uncertain information [1,2,3,4]. In the 1970s and 1980s cognitive scientists generally came to the view that these characteristic failures come about because people revise their beliefs by processes that bear little resemblance to formal mathematical calculation. Kahneman and Tversky developed a S. Quaglini, P. Barahona, and S. Andreassen (Eds.): AIME 2001, LNAI 2101, pp , Springer-Verlag Berlin Heidelberg 2001
2 Quantitative and Qualitative Approaches to Reasoning under Uncertainty 273 celebrated account of human decision-making and its weaknesses in terms of what they called heuristics and biases [5]. They argued that people judge things to be highly likely when, for example, they come easily to mind or are typical of a class rather than by means of a proper calculation of the relative probabilities. Such heuristic methods are often reasonable approximations for practical decision making but they can also lead to systematic errors. If people demonstrate imperfect reasoning or decision-making then it would presumably be desirable to support them with techniques that avoid errors and comply with rational rules. Mathematical logic (notably "classical" propositional logic and the predicate calculus) is traditionally taken as the gold standard for logical reasoning and deduction, while expected utility theory (EUT) plays the equivalent role for decisionmaking. A standard view on the "correct" way to take decisions is summarised by Lindley as follows: "... there is essentially only one way to reach a decision sensibly. First, the uncertainties present in the situation must be quantified in terms of values called probabilities. Second, the consequences of the courses of actions must be similarly described in terms of utilities. Third, that decision must be taken which is expected on the basis of the calculated probabilities to give the greatest utility. The force of must used in three places there is simply that any deviation from the precepts is liable to lead the decision maker in procedures which are demonstrably absurd" [6], p.vii. However, many people think that this overstates the value of mathematical methods and understates the capabilities of human decision makers. There are a number of problems with EUT as a practical method for decision making, and there are indications that, far from being "irrational", human decision processes depart from EUT because they are optimised to make various tradeoffs to address these problems in practical situations. An expected-utility decision procedure requires that we know, or can estimate reasonably accurately, all the required probability and utility parameters. This is frequently difficult in real-world situations since a decision may still be urgently required even if precise quantitative data are not available. Even when it is possible to establish the necessary parameters, the cost of obtaining good estimates may outweigh the expected benefits. Furthermore, in many situations a decision is needed before the decision options, or the relevant information sources, are fully known. The complete set of options may only emerge as the decision making process evolves. The potential value of mathematical decision theory is frequently limited by the lack of objective quantitative data on which to base the calculations, the limited range of functions that it can be used to support, and the problem that the underlying numerical representation of the decision is very different from the intuitive understanding of human decision makers. Human decision-making may also not be as "absurd" as the normative theories appear to suggest. One school of thought argues that many apparent biases and shortcomings are actually artefacts of the highly artificial situations that researchers create in order to study reasoning and judgement in controlled laboratory conditions. When we look at real-world decision-making we see that human reasoning and decision-making is more impressive than the research implies.
3 274 J. Fox, D. Glasspool, and J. Bury Shanteau [7] studied expert decision-making, investigating factors that lead to competence in experts, as opposed to the usual emphasis on incompetence". He identified a number of important positive characteristics of expert decision-makers: First, they know what is relevant to specific decisions, what to attend to in a busy environment, and they know when to make exceptions to general rules 1. Secondly, experts know a lot about what they know, and can make decisions about their own decision processes: they know which decisions to make and when, and which to skip, for example. They often have good communication skills and the ability to articulate their decisions and how they arrived at them. They can adapt to changing task conditions, and are frequently able to find novel solutions to problems. Classical deduction and probabilistic reasoning do not deal with these meta-cognitive skills. It has also been strongly argued that people are well adapted for making decisions under adverse conditions: time pressure, lack of detailed information and knowledge etc. Gigerenzer, for instance, has suggested that people make decisions in a "fast and frugal" way, which is to say human cognition is rational in the sense that it is optimised for speed at the cost of occasional and usually inconsequential errors [4]. 2.1 Tradeoffs in Effective Decision-Making The strong view expressed by Lindley and others is that the only rational or "coherent" way to reason with uncertainty is to require that we comply with certain mathematical axioms - the axioms of EUT. In practice compliance with these axioms is often difficult because there is insufficient data to permit a valid calculation of the expected utility of the decision options, or a valid calculation may require too much time. An alternative perspective is that human decision-making is rational in that it incorporates practical tradeoffs that, for example, trade a lower cost (e.g. errors in decision-making) against a higher one (e.g. the amount of input data required or the time taken to calculate expected utility). Tradeoffs of this kind not only simplify decision-making but in practice may entail only modest costs in the accuracy or effectiveness of decision-making. Consequently the claim that we should not model decision-support systems on human cognitive processes is less compelling than it may at first appear. This possibility has been studied extensively in the field of medical decisionmaking. In the prediction of sudden infant death, for example, Carpenter et al. [8] attempted to predict death from a simple linear combination of eight variables. They found that weights can be varied across a broad range without decreasing predictive accuracy. In diagnosing patients suffering from dyspepsia, Fox et al. [9] found that giving all pieces of evidence equal weights produced the same accuracy as a more precise statistical method (and also much the same pattern of errors). In another study [10] we developed a system for the interpretation of blood data in leukaemia diagnosis, using the EMYCIN expert system shell. EMYCIN provided facilities to attach numerical "certainty factors" to inference rules. Initially a system was developed 1 Good surgeons, the saying goes, know how to operate, better surgeons know when to operate and the best surgeons know when not to operate. That's true for all of medicine Richard Smith, Editor of the British Medical Journal.
4 Quantitative and Qualitative Approaches to Reasoning under Uncertainty 275 using the full range of available values (-1 to +1). These values were then replaced with just two: if the rule made a purely categorical inference the certainty factor was set to be 1.0 while if there was any uncertainty associated with the rule the certainty factor was set to 0.5. The effect was to increase diagnostic accuracy by 5%! In a study of whether or not to admit patients with suspected heart attacks to hospital by O Neil and Glowinski [11] no advantage was found of a precise decision procedure over simply "adding up the pros and cons". A similar comparison by Pradhan et al in a diagnosis task [12] showed a slight increase in accuracy of diagnosis with precise statistical reasoning, but the effect was so small it would have no practical clinical value. While the available evidence is not conclusive, a provisional hypothesis is that for some decisions strict use of quantitatively precise decision-making methods may not add much practical value to the design of decision support systems over simpler, more ad hoc methods. 3 Qualitative Methods for Decision Making Work in artificial intelligence raises an even more radical option for the designers of decision support systems. The desire to develop versatile automata has stimulated a great deal of research in new methods of decision making under uncertainty, ranging from sophisticated refinements of probabilistic methods such as Bayesian networks and Dempster-Shafer belief functions to non-probabilistic methods such as fuzzy logic and possibility theory. Good overviews of the different approaches and their applications are [13, 14]. These "non-standard" approaches are similar to probability methods in that they treat uncertainty as a matter of degree. However, even this apparently innocuous assumption has also been widely questioned on the practical grounds that they also demand a great deal of quantitative input data and also that decision-makers often find them difficult to understand because they do not capture intuitions about uncertainty - the nature of "belief", "doubt" and the form of natural justifications for decisionmaking. Consequently, interest has grown in AI in the use of non-numerical methods that seem to have some "common-sense" validity for reasoning under uncertainty but are not ad hoc from a formal point of view. These include non-classical logics including non-monotonic logics, default logic and defeasible reasoning. Cognitive approaches, sometimes called reason-based decision making, are also gaining ground, including the idea of using informal endorsements for alternative decision options [15] and formalisations of everyday strategies of reasoning about competing beliefs and actions based on logical arguments [16, 17, 18] Argumentation is a formalisation of the idea that decisions are made on the basis of arguments for or against a claim. Fox and Das [19] propose that argumentation may be the basis of a generalised decision theory, embracing standard probability theory as a special case, as well as other qualitative and semi-quantitative approaches. To take an example, suppose we wished to make the following informal argument:
5 276 J. Fox, D. Glasspool, and J. Bury If three or more first degree relatives of a patient have contracted breast cancer, then this is one reason to believe that the patient carries a gene predisposing to breast cancer. In the scheme of [19] arguments are defined as logical structures having three terms: (Claim, Grounds, Qualifier) In the example the Claim term would be the proposition "this patient carries a gene predisposing to breast cancer" and the Grounds "three or more first degree relatives of this patient have contracted breast cancer" is the justification for the argument. The final term, Qualifier, specifies the nature and strength of the argument which can be drawn from the grounds to the claim. In the example the qualifier is informal "this is one reason to believe", but it could be more conventional, as in "given the Grounds are true the Claim is true with a conditional probability of 0.43". Qualifiers are specified with reference to a particular dictionary of terms, along with an aggregation function that specifies how qualifiers in multiple arguments are to be combined. One possible dictionary is the set of real numbers from 0.0 to 1.0, which would allow qualifiers to be specified as precise numerical probability values and, with an appropriate aggregation function based on the theorems of classical probability, would allow the scheme to reduce to standard probability theory. In many situations little can be known of the strength of arguments, other than that they indicate an increase or decrease in the overall confidence in the claim. In this case the dictionary and aggregation function might be simpler. In [20] for example we describe a system which assesses carcinogenicity of novel compounds using a dictionary comprising the symbols + (argument in favour), - (argument against). The aggregation function in this case simply sums the number of arguments in favour of a proposition minus those against. Other dictionaries can include additional qualifiers, such as ++ (the argument confirms the claim) and -- (the argument refutes the claim). Another possible dictionary adopts linguistic confidence terms (e.g. probable, improbable, possible, plausible etc.) and requires a logical aggregation function that provides a formalised semantics for combining such terms according to common usage. Such terms have been formally categorised by their logical structure for example [21, 22]. A feature of argumentation theory is that it subsumes these apparently diverse approaches within a single formalism which has a clear consistent formal semantics [19]. Additionally, argumentation gives a different perspective on the decision making process which we believe to be more in line with the way people naturally think about probability and possibility than standard probability theory. Given the evidence reviewed above that people do not naturally use EUT in their everyday decision making, our tentative position is that expressing a preference or confidence in terms of arguments for and against each option will be more accessible and comprehensible than providing a single number representing its aggregate probability. We have developed a number of decision support systems to explore this idea. In the next sections we describe a technology which supports an argument-based decision procedure, then outline several applications which have been built using it, and finally consider quantitative evaluations of the applications.
6 Quantitative and Qualitative Approaches to Reasoning under Uncertainty Practical Applications of the Argumentation Method We have used the argumentation framework in designing five decision support systems to date. 4.1 CAPSULE The CAPSULE (Computer Aided Prescribing Using Logic Engineering) system was developed to assist GPs with prescribing decisions [19,23]. CAPSULE analyses patient notes and constructs a list of relevant candidate medications, together with arguments for and against each option (based on nine different criteria, including efficacy, contra-indications, drug interactions, side effects, costs etc). 4.2 CADMIUM The CADMIUM radiology workstation is an experimental package for combining image processing with logic-based decision support. The main use that CADMIUM has been applied to is in screening for breast cancer, in which an argument based decision procedure has the dual function of controlling image processing functions which extract and describe micro-calcifications in breast x-rays and interprets the descriptions in terms of whether they are likely to indicate benign or malignant abnormalities. The decision procedure assesses the arguments and presents them to the user in a structured report. 4.3 RAGs The RAGs (Risk Assessment in Genetics) system allows the user to describe a patient s family tree incorporating information on the known incidence of cancers within the family. This information is evaluated to assess the likelihood that a genetic predisposition to a particular cancer is present. The software makes recommendations for patient managementin language which is comprehensible to both clinician and patient. A family tree graphic is used for incremental data entry (Figure 1). Data about relatives are added by clicking on each relative s icon, and completing a simple form. RAGs analyses the data and provides detailed assessments of genetic risk by weighing simple arguments like the example above rather than using numerical probabilities. Based on the aggregate genetic risk level the patient is classified as at high, moderate or low risk of carrying a BrCa1 genetic mutation, which implies an 80% lifetime risk of developing breast cancer. Appropriate referral advice can be given for the patient based on this classification. The software can provide a comprehensible explanation for its decisions based on the arguments it has applied (Figure 1). RAGs uses a set of 23 risk arguments (e.g. if the client has more than two firstdegree relatives with breast cancer under the age of 50 then this is a risk factor) and a simple dictionary of qualifiers which allows small positive and negative integer values as well as plus and minus infinity (equivalent to confirming or refuting the
7 278 J. Fox, D. Glasspool, and J. Bury claim). This scheme thus preserves relative rather than absolute weighting of different factors in the analysis. 4.4 ERA Patients with symptoms or signs which may indicate cancer should be investigated quickly so that treatment may commence as soon as possible. The ERA (Early Referrals Application) system has been designed in the context of the UK Department of Health s (DoH) 2 week guideline which states that patients with suspected cancer should be seen and assessed by an appropriate specialist within 2 week of their presentation. Referral criteria are given for each of 12 main cancer groups (e.g. Breast, Lung, Colorectal). The practical decision as to whether or not to refer a particular patient is treated in ERA as based on a set of patient-specific arguments (see figure 2). The published referral criteria have been specifically designed to be unambiguous and easy to apply. Should any of the arguments apply to a particular patient, an early referral is warranted. Qualifiers as to the weight of each argument are thus unnecessary, with arguments generally acting categorically in an all or nothing fashion. An additional feature of this domain is that there are essentially no counter-arguments. The value of argumentation in this example lies largely in the ability to give meaningful explanations in the form of reasons for referral expressed in English. 4.5 REACT The applications described so far involve making a single decision - what drug to prescribe, or whether to refer a patient. Most decisions are made in the context of plans of action, however, where they may interact or conflict with other planned actions or anticipated events. REACT (Risk, Events, Actions and their Consequences over Time) is being developed to provide decision support for extended plans. In effect REACT is a logical spreadsheet that allows a user to manipulate graphical widgets representing possible clinical events and interventions on a timeline interface and propagates their implications (both qualitative and quantitative) to numerical displays of risk (or other parameters) and displays of arguments and counterarguments. While the REACT user creates a plan, a knowledge-based DSS analyses it according to a set of definable rules and provides feedback on interactions between events. Rules may specify, for example, that certain events are mutually exclusive, that certain combinations of events are impossible, or that events have different consequences depending on prior or simultaneous events). Global measures (for example the predicted degree of risk or predicted cost or benefit of combinations of events) can be displayed graphically alongside the planning timeline. Qualitative arguments for and against each individual action proposed in the plan can be reviewed, and can be used to generate recommended actions when specified combinations of plan elements occur.
8 Quantitative and Qualitative Approaches to Reasoning under Uncertainty 279 Fig. 1. The RAGs software application. A family tree has been input for the patient Karen, who has had three relatives affected by relevant cancers. The risk assessment system has determined a pattern of inheritance that may indicate a genetic factor, and provides an explanation in the left-hand panel. Referral advice for the patient is also available. Fig. 2. The ERA early referral application. After completing patient details on a form, the program provides referral recommendations with advice, and can contact the hospital to automatically make an appointment.
9 280 J. Fox, D. Glasspool, and J. Bury 5 Evaluations A number of these applications have been used to carry out controlled evaluations (with the exception of ERA and REACT which are still in development. In the case of the CAPSULE prescribing system a controlled study with Oxfordshire general practitioners showed the potential for substantial improvements in the quality of their prescribing decisions [23]. With decision support there was a 70% increase in the number of times the GPs decisions agreed with those of experts considering the cases, and a 50% reduction in the number of times that they missed a cheaper but equally effective medication. The risk classifications generated by the RAGS software was compared for 50 families with that provided by the leading probabilistic risk assessment software. This uses the Claus model [27], a mathematical model of genetic risk of breast cancer based on a large dataset of cancer cases. Despite the use of a very simple weighting scheme the RAGs system produced exactly the same risk classification (into high, medium or low risk, according to established guidelines) for all cases as the probabilistic system [24, 25]. RAGs resulted in more accurate pedigree taking and more appropriate management decisions than either pencil and paper or standard probabilistic software [26]. In the CADMIUM image-processing system users are provided with assistance in interpreting mammograms using an argument-based decision making component. Radiographers who were trained to interpret mammograms were asked to make decisions as to whether observed abnormalities were benign or malignant, with and without decision support. The decision support condition produced clear improvements in the radiographers performance, in terms of increased hits and correct rejections and reduced misses and false positives [27]. 6 Summary and Conclusions Human reasoning and decision-making can exhibit various shortcomings when compared with accepted prescriptive theories. Good decision-making is central to many important human activities and much effort is directed at developing decision support technologies to assist us. One school of thought argues that these systems must be based on prescriptive axioms of decision-making since to do otherwise leads inevitably to irrational conclusions and choices. Others argue that decision-making in the real world demands practical tradeoffs and a failure to make those tradeoffs would be irrational. Although the debate remains inconclusive, we have described a number of examples of medical systems in which the use of simple logical argumentation appears to provide effective decision support together with a versatile representation of uncertainty which fits comfortably with people's intuitions. Acknowledgements. The RAGs project was supported by the Economic and Social Research Council, and much of the development and evaluation work was carried out by Andrew Coulson and Jon Emery. Michael Humber carried out much of the implementation of the ERA system.
10 References Quantitative and Qualitative Approaches to Reasoning under Uncertainty Kahneman D., Slovic P. and Tversky A. (eds): Heuristics and Biases. Cambridge University Press, Cambridge (1982) 2. Evans J. st B. and Over D.E.: Rationality and reasoning. Psychology press, London (1996) 3. Wright G. and Ayton P. (eds): Subjective Probability. Wiley, Chichester, UK (1994) 4. Gigerenzer G. and Todd P.M.: Simple heuristics that make us smart. Oxford University Press, Oxford (1999) 5. Kahneman, D. and Tversky, A.: Judgement under Uncertainty: Heuristics and Biases. Science, Vol. 185, 27 (1974) pp Lindley DV.: Making Decisions (2 nd Edition). Wiley, Chichester, UK (1985) 7. Shanteau, J.: Psychological characteristics of expert decision makers. In, J Mumpower (ed) Expert judgement and expert systems. NATO ASI Series, volume F35 (1987) 8. Carpenter R G, Garnder A, McWeeny P M. and Emery J L.: Multistage scoring system for identifying infants at risk of unexpected death. Archives of disease in childhood, 53(8), (1977) pp Fox J. Barber DC. and Bardhan KD.: Alternatives to Bayes? A quantitative comparison with rule-based diagnosis. Methods of Information in Medicine, 10 (4) (1980) pp Fox, J. Myers CD., Greaves MF. and Pegram S.: Knowledge acquisition for expert systems: experience in leukaemia diagnosis. Methods of Information in Medicine, 24 (1) (1985) pp O Neil MJ. and Glowinski AJ.: Evaluating and validating very large knowledge-based systems. Medical Informatics, 15 (3) (1990) pp Pradhan M.: The sensitivity of belief networks to imprecise probabilities: an experimental investigation. Artificial Intelligence Journal, 84 (1-2) (1996) pp Krause P. and Clark C.: Uncertainty and Subjective Probability in AI Systems In Wright G & Ayton P (Eds) Subjective Probability, Wiley J & Sons (1994) pp Hunter, A. and Parsons, S. (Editors) Applications of uncertainty formalisms, Springer Verlag, LNAI 1455, (1998) 15. Cohen P.R.: Heuristic Reasoning: An Artificial Intelligence Approach. Pitman Advanced Publishing Program, Boston (1985) 16. Fox J.: On the necessity of probability: Reasons to believe and grounds for doubt. In Wright G, Ayton, P, eds., Subjective Probability. John Wiley, Chichester (1994) 17. Fox J, Krause P. and Ambler S.: Arguments, contradictions and practical reasoning. In Neumann B, ed. Proceedings of the 10 th European Conference on AI (ECAI92), Vienna, Austria (1992) pp Curley SP. and Benson PG.: Applying a cognitive perspective to probability construction. In G Wright & P Ayton (eds.), Subjective Probability. John Wiley & Sons. Chichester, England (1994) pp Fox J. and Das S K.: Safe and Sound: Artificial Intelligence in Hazardous Applications, American Association of Artificial Intelligence and MIT Press (2000) 20. Tonnelier CAG., Fox J., Judson P., Krause P., Pappas N. and Patel M.: Representation of Chemical Structures in Knowledge-Based Systems. J. Chem. Inf. Sci. 37 (1997) pp Elvang-Goransson M, Krause P.J and Fox J.: Acceptability of Arguments as Logical Uncertainty. In: Clarke M, Kruse R and Moral S, eds. Symbolic and Quantitative Approaches to Reasoning and Uncertainty. Proceedings of European Conference ECSQARU93. Lecture Notes in Computer Science 747. Springer-Verlag (1993) pp Glasspool DW. and Fox J.: Understanding probability words by constructing concrete mental models. In Hahn M, Stoness, SC, eds, Proceedings of the 21 st Annual Conference of the Cognitive Science Society (1999) pp
11 282 J. Fox, D. Glasspool, and J. Bury 23. Walton RT, Gierl C, Yudkin P, Mistry H, Vessey MP & Fox J.: Evaluation of computer support for prescribing (CAPSULE) using simulated cases. British Medical Journal 315 (1997) pp Emery J, Watson E, Rose P and Andermann, A. A systematic review of the literature exploring the role of primary care in genetic services. Fam. Prac. 16 (1999) pp Coulson AS, Glasspool DW, Fox J and Emery J. Computerized Genetic Risk Assessment from Family Pedigrees. MD computing, in press. 26. Emery J., Walton, R., Murphy, M., Austoker, J., Yudkin, P., Chapman, C., Coulson, A., Glasspool, D. and Fox, J.: Computer support for interpreting family histories of breast and ovarian cancer in primary care: comparative study with simulated cases. British Medical Journal, 321 (2000) pp Claus E, Schildkraut J, Thompson WD and Risch, N. The genetic attributable risk of breast and ovarian cancer. Cancer 1996; 77, pp Taylor, P, Fox J and Todd-Pokropek, A The development and evaluation of CADMIUM: a prototype system to assist in the interpretation of mammograms Medical Image Analysis, 1999, 3 (4),
Towards a general model for argumentation services
Towards a general model for argumentation services John Fox 1,2, Liz Black 1, David Glasspool 1,2, Sanjay Modgil 1,2, Ayelet Oettinger 1, Vivek Patkar 1, Matt Williams 1 1 Advanced Computation Laboratory,
More informationDecision support for health care: the PROforma evidence base
Informatics in Primary Care 2006;14:49 54 # 2006 PHCSG, British Computer Society Conference paper Decision support for health care: the PROforma evidence base John Fox PhD Head of Laboratory Vivek Patkar
More informationMedical thinking: towards a unified view. John Fox Advanced Computation Laboratory Cancer Research UK
Medical thinking: towards a unified view John Fox Advanced Computation Laboratory Cancer Research UK Cognitive theory Clinical The CREDO trial PROforma process modelling trials Theory to practice Experiments
More informationWason's Cards: What is Wrong?
Wason's Cards: What is Wrong? Pei Wang Computer and Information Sciences, Temple University This paper proposes a new interpretation
More informationArtificial Intelligence and Human Thinking. Robert Kowalski Imperial College London
Artificial Intelligence and Human Thinking Robert Kowalski Imperial College London 1 Artificial Intelligence and Human Thinking The Abductive Logic Programming (ALP) agent model as a unifying framework
More informationVisual book review 1 Safe and Sound, AI in hazardous applications by John Fox and Subrata Das
Visual book review 1 Safe and Sound, AI in hazardous applications by John Fox and Subrata Das Boris Kovalerchuk Dept. of Computer Science Central Washington University, Ellensburg, WA 98926-7520 borisk@cwu.edu
More informationKnowledge is rarely absolutely certain. In expert systems we need a way to say that something is probably but not necessarily true.
CmSc310 Artificial Intelligence Expert Systems II 1. Reasoning under uncertainty Knowledge is rarely absolutely certain. In expert systems we need a way to say that something is probably but not necessarily
More informationConfirmation Bias. this entry appeared in pp of in M. Kattan (Ed.), The Encyclopedia of Medical Decision Making.
Confirmation Bias Jonathan D Nelson^ and Craig R M McKenzie + this entry appeared in pp. 167-171 of in M. Kattan (Ed.), The Encyclopedia of Medical Decision Making. London, UK: Sage the full Encyclopedia
More informationHow Does Analysis of Competing Hypotheses (ACH) Improve Intelligence Analysis?
How Does Analysis of Competing Hypotheses (ACH) Improve Intelligence Analysis? Richards J. Heuer, Jr. Version 1.2, October 16, 2005 This document is from a collection of works by Richards J. Heuer, Jr.
More informationSupplementary notes for lecture 8: Computational modeling of cognitive development
Supplementary notes for lecture 8: Computational modeling of cognitive development Slide 1 Why computational modeling is important for studying cognitive development. Let s think about how to study the
More informationBayesian Networks in Medicine: a Model-based Approach to Medical Decision Making
Bayesian Networks in Medicine: a Model-based Approach to Medical Decision Making Peter Lucas Department of Computing Science University of Aberdeen Scotland, UK plucas@csd.abdn.ac.uk Abstract Bayesian
More information[1] provides a philosophical introduction to the subject. Simon [21] discusses numerous topics in economics; see [2] for a broad economic survey.
Draft of an article to appear in The MIT Encyclopedia of the Cognitive Sciences (Rob Wilson and Frank Kiel, editors), Cambridge, Massachusetts: MIT Press, 1997. Copyright c 1997 Jon Doyle. All rights reserved
More informationTEACHING YOUNG GROWNUPS HOW TO USE BAYESIAN NETWORKS.
TEACHING YOUNG GROWNUPS HOW TO USE BAYESIAN NETWORKS Stefan Krauss 1, Georg Bruckmaier 1 and Laura Martignon 2 1 Institute of Mathematics and Mathematics Education, University of Regensburg, Germany 2
More informationStepwise Knowledge Acquisition in a Fuzzy Knowledge Representation Framework
Stepwise Knowledge Acquisition in a Fuzzy Knowledge Representation Framework Thomas E. Rothenfluh 1, Karl Bögl 2, and Klaus-Peter Adlassnig 2 1 Department of Psychology University of Zurich, Zürichbergstraße
More informationCISC453 Winter Probabilistic Reasoning Part B: AIMA3e Ch
CISC453 Winter 2010 Probabilistic Reasoning Part B: AIMA3e Ch 14.5-14.8 Overview 2 a roundup of approaches from AIMA3e 14.5-14.8 14.5 a survey of approximate methods alternatives to the direct computing
More informationEvaluation Models STUDIES OF DIAGNOSTIC EFFICIENCY
2. Evaluation Model 2 Evaluation Models To understand the strengths and weaknesses of evaluation, one must keep in mind its fundamental purpose: to inform those who make decisions. The inferences drawn
More informationModels for Inexact Reasoning. Imprecision and Approximate Reasoning. Miguel García Remesal Department of Artificial Intelligence
Models for Inexact Reasoning Introduction to Uncertainty, Imprecision and Approximate Reasoning Miguel García Remesal Department of Artificial Intelligence mgremesal@fi.upm.es Uncertainty and Imprecision
More informationJanuary 2, Overview
American Statistical Association Position on Statistical Statements for Forensic Evidence Presented under the guidance of the ASA Forensic Science Advisory Committee * January 2, 2019 Overview The American
More informationKahneman, Daniel. Thinking Fast and Slow. New York: Farrar, Straus & Giroux, 2011.
The accumulating research indicates that individuals cognitive and behavioral orientations to objects (their thoughts and actions) are frequently based on rapid shortcuts or heuristics. The past few decades
More informationRepresentation Theorems for Multiple Belief Changes
Representation Theorems for Multiple Belief Changes Dongmo Zhang 1,2 Shifu Chen 1 Wujia Zhu 1,2 Zhaoqian Chen 1 1 State Key Lab. for Novel Software Technology Department of Computer Science and Technology
More informationInformation in practice
Computer support for interpreting family histories of breast and ovarian cancer in primary care: comparative study with simulated cases Jon Emery, Robert Walton, Michael Murphy, Joan Austoker, Pat Yudkin,
More informationArtificial Intelligence Programming Probability
Artificial Intelligence Programming Probability Chris Brooks Department of Computer Science University of San Francisco Department of Computer Science University of San Francisco p.1/25 17-0: Uncertainty
More informationImproving statistical estimates used in the courtroom. Precis. Bayes Theorem. Professor Norman Fenton. Queen Mary University of London and Agena Ltd
Improving statistical estimates used in the courtroom Professor Norman Fenton Queen Mary University of London and Agena Ltd Address: Queen Mary University of London School of Electronic Engineering and
More informationIntroduction to Belief Functions 1
Introduction to Belief Functions 1 Rajendra P. Srivastava* and Theodore J. Mock** *University of Kansas, **University of Southern California 1. Introduction The refusal to choose is a form of choice; Disbelief
More informationQuantifying uncertainty under a predictive, epistemic approach to risk analysis
Quantifying uncertainty under a predictive, epistemic approach to risk analysis S.Apeland Stavanger University College, Norway T.Aven Stavanger University College, Norway T.Nilsen RF Rogaland Research,
More informationThe Role of Causality in Judgment Under Uncertainty. Tevye R. Krynski & Joshua B. Tenenbaum
Causality in Judgment 1 Running head: CAUSALITY IN JUDGMENT The Role of Causality in Judgment Under Uncertainty Tevye R. Krynski & Joshua B. Tenenbaum Department of Brain & Cognitive Sciences, Massachusetts
More informationCitation for published version (APA): Geus, A. F. D., & Rotterdam, E. P. (1992). Decision support in aneastehesia s.n.
University of Groningen Decision support in aneastehesia Geus, Arian Fred de; Rotterdam, Ernest Peter IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to
More informationStatistics and Full Belief
Commentary on Descartes Lectures 2014 Statistics and Full Belief Or: How to Judge if You Must Jan-Willem Romeijn University of Groningen Joint work with Hannes Leitgeb LMU Munich Ring-ring You re called
More informationThe Logic of Categorization
From: FLAIRS-02 Proceedings. Copyright 2002, AAAI (www.aaai.org). All rights reserved. The Logic of Categorization Pei Wang Department of Computer and Information Sciences, Temple University Philadelphia,
More informationIrrationality in Game Theory
Irrationality in Game Theory Yamin Htun Dec 9, 2005 Abstract The concepts in game theory have been evolving in such a way that existing theories are recasted to apply to problems that previously appeared
More informationBayesian Belief Network Based Fault Diagnosis in Automotive Electronic Systems
Bayesian Belief Network Based Fault Diagnosis in Automotive Electronic Systems Yingping Huang *, David Antory, R. Peter Jones, Craig Groom, Ross McMurran, Peter Earp and Francis Mckinney International
More informationTHE USE OF MULTIVARIATE ANALYSIS IN DEVELOPMENT THEORY: A CRITIQUE OF THE APPROACH ADOPTED BY ADELMAN AND MORRIS A. C. RAYNER
THE USE OF MULTIVARIATE ANALYSIS IN DEVELOPMENT THEORY: A CRITIQUE OF THE APPROACH ADOPTED BY ADELMAN AND MORRIS A. C. RAYNER Introduction, 639. Factor analysis, 639. Discriminant analysis, 644. INTRODUCTION
More informationThinking and Intelligence
Thinking and Intelligence Learning objectives.1 The basic elements of thought.2 Whether the language you speak affects the way you think.3 How subconscious thinking, nonconscious thinking, and mindlessness
More informationA Unified View of Consequence Relation, Belief Revision and Conditional Logic
A Unified View of Consequence Relation, Belief Revision and Conditional Logic Hirofumi Katsuno NTT Basic Research Laboratories Musashinoshi, Tokyo 180 Japan katsuno(dntt-20.ntt. jp Ken Satoh ICOT Research
More informationA FRAMEWORK FOR CLINICAL DECISION SUPPORT IN INTERNAL MEDICINE A PRELIMINARY VIEW Kopecky D 1, Adlassnig K-P 1
A FRAMEWORK FOR CLINICAL DECISION SUPPORT IN INTERNAL MEDICINE A PRELIMINARY VIEW Kopecky D 1, Adlassnig K-P 1 Abstract MedFrame provides a medical institution with a set of software tools for developing
More informationMTAT Bayesian Networks. Introductory Lecture. Sven Laur University of Tartu
MTAT.05.113 Bayesian Networks Introductory Lecture Sven Laur University of Tartu Motivation Probability calculus can be viewed as an extension of classical logic. We use many imprecise and heuristic rules
More informationChapter 11 Decision Making. Syllogism. The Logic
Chapter 11 Decision Making Syllogism All men are mortal. (major premise) Socrates is a man. (minor premise) (therefore) Socrates is mortal. (conclusion) The Logic Mortal Socrates Men 1 An Abstract Syllogism
More informationAnswers to end of chapter questions
Answers to end of chapter questions Chapter 1 What are the three most important characteristics of QCA as a method of data analysis? QCA is (1) systematic, (2) flexible, and (3) it reduces data. What are
More informationAgenetic disorder serious, perhaps fatal without
ACADEMIA AND CLINIC The First Positive: Computing Positive Predictive Value at the Extremes James E. Smith, PhD; Robert L. Winkler, PhD; and Dennis G. Fryback, PhD Computing the positive predictive value
More informationKevin Burns. Introduction. Foundation
From: AAAI Technical Report FS-02-01. Compilation copyright 2002, AAAI (www.aaai.org). All rights reserved. Dealing with TRACS: The Game of Confidence and Consequence Kevin Burns The MITRE Corporation
More informationCosting report: Lipid modification Implementing the NICE guideline on lipid modification (CG181)
Putting NICE guidance into practice Costing report: Lipid modification Implementing the NICE guideline on lipid modification (CG181) Published: July 2014 This costing report accompanies Lipid modification:
More informationModule 14: Missing Data Concepts
Module 14: Missing Data Concepts Jonathan Bartlett & James Carpenter London School of Hygiene & Tropical Medicine Supported by ESRC grant RES 189-25-0103 and MRC grant G0900724 Pre-requisites Module 3
More informationINTRODUCTION TO BAYESIAN REASONING
International Journal of Technology Assessment in Health Care, 17:1 (2001), 9 16. Copyright c 2001 Cambridge University Press. Printed in the U.S.A. INTRODUCTION TO BAYESIAN REASONING John Hornberger Roche
More informationStructural assessment of heritage buildings
Defence Sites 69 Structural assessment of heritage buildings M. Holicky & M. Sykora Klokner Institute, Czech Technical University in Prague, Czech Republic Abstract Reliability assessment of heritage buildings
More informationRELYING ON TRUST TO FIND RELIABLE INFORMATION. Keywords Reputation; recommendation; trust; information retrieval; open distributed systems.
Abstract RELYING ON TRUST TO FIND RELIABLE INFORMATION Alfarez Abdul-Rahman and Stephen Hailes Department of Computer Science, University College London Gower Street, London WC1E 6BT, United Kingdom E-mail:
More informationWhat Is A Knowledge Representation? Lecture 13
What Is A Knowledge Representation? 6.871 - Lecture 13 Outline What Is A Representation? Five Roles What Should A Representation Be? What Consequences Does This View Have For Research And Practice? One
More informationCan Bayesian models have normative pull on human reasoners?
Can Bayesian models have normative pull on human reasoners? Frank Zenker 1,2,3 1 Lund University, Department of Philosophy & Cognitive Science, LUX, Box 192, 22100 Lund, Sweden 2 Universität Konstanz,
More informationFinding Information Sources by Model Sharing in Open Multi-Agent Systems 1
Finding Information Sources by Model Sharing in Open Multi-Agent Systems Jisun Park, K. Suzanne Barber The Laboratory for Intelligent Processes and Systems The University of Texas at Austin 20 E. 24 th
More informationOn some misconceptions about subjective probability and Bayesian inference Coolen, F.P.A.
On some misconceptions about subjective probability and Bayesian inference Coolen, F.P.A. Published: 01/01/1991 Document Version Publisher s PDF, also known as Version of Record (includes final page, issue
More informationCambridge International AS & A Level Global Perspectives and Research. Component 4
Cambridge International AS & A Level Global Perspectives and Research 9239 Component 4 In order to help us develop the highest quality Curriculum Support resources, we are undertaking a continuous programme
More informationCOGNITIVE BIAS IN PROFESSIONAL JUDGMENT
COGNITIVE BIAS IN PROFESSIONAL JUDGMENT Dr. Janet L Sutton, PhD Behavioral Economist, Senior Research Psychologist U.S. Department of Defense (retired): -Air Force Research Laboratory (AFRL, Wright-Patterson
More informationDRAFT. full paper published in: Lecture Notes in Computer Science, Volume 5219, Berlin/Heidelberg, 2008, pp
DRAFT full paper published in: Lecture Notes in Computer Science, Volume 5219, Berlin/Heidelberg, 2008, pp. 291-304 Proceedings of the 27 th International Conference on Computer Safety, Reliability and
More informationEEL-5840 Elements of {Artificial} Machine Intelligence
Menu Introduction Syllabus Grading: Last 2 Yrs Class Average 3.55; {3.7 Fall 2012 w/24 students & 3.45 Fall 2013} General Comments Copyright Dr. A. Antonio Arroyo Page 2 vs. Artificial Intelligence? DEF:
More informationSTRATEGIC COST ADVICE AND THE EFFECT OF THE AVAILABILITY HEURISTIC C J FORTUNE
CIB World Building Congress, April 21, Wellington, New Zealand Page 1 of 9 STRATEGIC COST ADVICE AND THE EFFECT OF THE AVAILABILITY HEURISTIC C J FORTUNE School of the Built Environment, Liverpool John
More informationThe Common Priors Assumption: A comment on Bargaining and the Nature of War
The Common Priors Assumption: A comment on Bargaining and the Nature of War Mark Fey Kristopher W. Ramsay June 10, 2005 Abstract In a recent article in the JCR, Smith and Stam (2004) call into question
More informationThe role of theory in construction management: a call for debate
The role of theory in construction management: a call for debate Seymour, D, Crook, D and Rooke, JA http://dx.doi.org/10.1080/014461997373169 Title Authors Type URL The role of theory in construction management:
More informationGenetic Disease, Genetic Testing and the Clinician
Clemson University TigerPrints Publications Philosophy & Religion 1-2001 Genetic Disease, Genetic Testing and the Clinician Kelly C. Smith Clemson University, kcs@clemson.edu Follow this and additional
More informationModelling Causation in Evidence-Based Legal Reasoning ICAIL Doctoral Consortium
Modelling Causation in Evidence-Based Legal Reasoning ICAIL Doctoral Consortium Ruta Liepina Law Department, European University Institute 1 Research Questions Research questions in this project revolve
More informationGeneralization and Theory-Building in Software Engineering Research
Generalization and Theory-Building in Software Engineering Research Magne Jørgensen, Dag Sjøberg Simula Research Laboratory {magne.jorgensen, dagsj}@simula.no Abstract The main purpose of this paper is
More informationThe idea of an essentially contested concept is incoherent.
Daniel Alexander Harris 1 The idea of an essentially contested concept is incoherent. Daniel Alexander Harris (2015) Daniel Alexander Harris 2 This essay will demonstrate the idea of an essentially contested
More informationExamples of Feedback Comments: How to use them to improve your report writing. Example 1: Compare and contrast
Examples of Feedback Comments: How to use them to improve your report writing This document contains 4 examples of writing and feedback comments from Level 2A lab reports, and 4 steps to help you apply
More informationWhy Does Similarity Correlate With Inductive Strength?
Why Does Similarity Correlate With Inductive Strength? Uri Hasson (uhasson@princeton.edu) Psychology Department, Princeton University Princeton, NJ 08540 USA Geoffrey P. Goodwin (ggoodwin@princeton.edu)
More informationOn the Representation of Nonmonotonic Relations in the Theory of Evidence
On the Representation of Nonmonotonic Relations in the Theory of Evidence Ronald R. Yager Machine Intelligence Institute Iona College New Rochelle, NY 10801 Abstract A Dempster-Shafer belief structure
More informationInferencing in Artificial Intelligence and Computational Linguistics
Inferencing in Artificial Intelligence and Computational Linguistics (http://www.dfki.de/~horacek/infer-ai-cl.html) no classes on 28.5., 18.6., 25.6. 2-3 extra lectures will be scheduled Helmut Horacek
More informationThe interplay of domain-specific and domain general processes, skills and abilities in the development of science knowledge
The interplay of domain-specific and domain general processes, skills and abilities in the development of science knowledge Stella Vosniadou Strategic Professor in Education The Flinders University of
More informationTechnical Specifications
Technical Specifications In order to provide summary information across a set of exercises, all tests must employ some form of scoring models. The most familiar of these scoring models is the one typically
More informationJournal of Experimental Psychology: Learning, Memory, and Cognition
Journal of Experimental Psychology: Learning, Memory, and Cognition Conflict and Bias in Heuristic Judgment Sudeep Bhatia Online First Publication, September 29, 2016. http://dx.doi.org/10.1037/xlm0000307
More informationThe Vagueness of Verbal Probability and Frequency Expressions
International Journal of Advanced Computer Science, Vol. 1, No. 2, Pp. 52-57, Aug. 2011. The Vagueness of Verbal Probability and Frequency Expressions Franziska Bocklisch Manuscript Received: 23, Aug.,
More informationAlcohol interventions in secondary and further education
National Institute for Health and Care Excellence Guideline version (Draft for Consultation) Alcohol interventions in secondary and further education NICE guideline: methods NICE guideline Methods
More informationReasoning with Uncertainty. Reasoning with Uncertainty. Bayes Rule. Often, we want to reason from observable information to unobservable information
Reasoning with Uncertainty Reasoning with Uncertainty Often, we want to reason from observable information to unobservable information We want to calculate how our prior beliefs change given new available
More informationRecognizing Ambiguity
Recognizing Ambiguity How Lack of Information Scares Us Mark Clements Columbia University I. Abstract In this paper, I will examine two different approaches to an experimental decision problem posed by
More informationReliability Analysis Combining Multiple Inspection Techniques
ECNDT 2006 - Fr.2.5.2 Reliability Analysis Combining Multiple Inspection Techniques Dag HORN, AECL Chalk River Laboratories, Chalk River, Ontario, Canada Abstract. The benefits of combining several independent
More informationUNESCO EOLSS. This article deals with risk-defusing behavior. It is argued that this forms a central part in decision processes.
RISK-DEFUSING BEHAVIOR Oswald Huber University of Fribourg, Switzerland Keywords: cognitive bias, control, cost of risk-defusing operators, decision making, effect of risk-defusing operators, lottery,
More informationExperimental Research in HCI. Alma Leora Culén University of Oslo, Department of Informatics, Design
Experimental Research in HCI Alma Leora Culén University of Oslo, Department of Informatics, Design almira@ifi.uio.no INF2260/4060 1 Oslo, 15/09/16 Review Method Methodology Research methods are simply
More informationOn Trust. Massimo Felici. Massimo Felici On Trust c
On Trust Massimo Felici Outline 1 What is Trust? Trust matters Trust problems Seeking an Experimental Trust Framework Trust and Dependability Directions for future work What is Trust? 2 Constructive Trust
More informationA Quantitative Model of Counterfactual Reasoning
A Quantitative Model of Counterfactual Reasoning Daniel Yarlett Division of Informatics University of Edinburgh Edinburgh, Scotland dany@cogsci.ed.ac.uk Michael Ramscar Division of Informatics University
More informationWho Is Rational? Studies of Individual Differences in Reasoning
Book Review/Compte rendu 291 Who Is Rational? Studies of Individual Differences in Reasoning Keith E. Stanovich Mahwah, NJ: Lawrence Erlbaum Associates, 1999. Pp. xvi, 1-296. Hardcover: ISBN 0-8058-2472-3,
More informationHybrid models of rational legal proof. Bart Verheij Institute of Artificial Intelligence and Cognitive Engineering
Hybrid models of rational legal proof Bart Verheij Institute of Artificial Intelligence and Cognitive Engineering www.ai.rug.nl/~verheij How can forensic evidence be handled effectively and safely? Analyses
More informationClinical Decision Support Systems. 朱爱玲 Medical Informatics Group 24 Jan,2003
Clinical Decision Support Systems 朱爱玲 Medical Informatics Group 24 Jan,2003 Outline Introduction Medical decision making What are clinical decision support systems (CDSSs)? Historical perspective Abdominal
More informationThe Role of Causal Models in Reasoning Under Uncertainty
The Role of Causal Models in Reasoning Under Uncertainty Tevye R. Krynski (tevye@mit.edu) Joshua B. Tenenbaum (jbt@mit.edu) Department of Brain & Cognitive Sciences, Massachusetts Institute of Technology
More informationMS&E 226: Small Data
MS&E 226: Small Data Lecture 10: Introduction to inference (v2) Ramesh Johari ramesh.johari@stanford.edu 1 / 17 What is inference? 2 / 17 Where did our data come from? Recall our sample is: Y, the vector
More informationBAYESIAN NETWORK FOR FAULT DIAGNOSIS
BAYESIAN NETWOK FO FAULT DIAGNOSIS C.H. Lo, Y.K. Wong and A.B. ad Department of Electrical Engineering, The Hong Kong Polytechnic University Hung Hom, Kowloon, Hong Kong Fax: +852 2330 544 Email: eechlo@inet.polyu.edu.hk,
More informationChapter 2. Knowledge Representation: Reasoning, Issues, and Acquisition. Teaching Notes
Chapter 2 Knowledge Representation: Reasoning, Issues, and Acquisition Teaching Notes This chapter explains how knowledge is represented in artificial intelligence. The topic may be launched by introducing
More informationA Belief-Based Account of Decision under Uncertainty. Craig R. Fox, Amos Tversky
A Belief-Based Account of Decision under Uncertainty Craig R. Fox, Amos Tversky Outline Problem Definition Decision under Uncertainty (classical Theory) Two-Stage Model Probability Judgment and Support
More informationThis is an Author's Original Manuscript of an article submitted to Behavior & Brain Science and may differ from the final version which is available here: http://journals.cambridge.org/action/displayabstract?frompage=online&aid=8242505
More informationArtificial Intelligence
Politecnico di Milano Artificial Intelligence Artificial Intelligence From intelligence to rationality? Viola Schiaffonati viola.schiaffonati@polimi.it Can machine think? 2 The birth of Artificial Intelligence
More informationPsychology as a Science of Design in Engineering
Proceedings of the British Psychological Society (vol. 7, No. 2) and the Bulletin of the Scottish Branch of the British Psychological Society in 1999. Psychology as a Science of Design in Engineering Patrik
More informationHow Different Choice Strategies Can Affect the Risk Elicitation Process
IAENG International Journal of Computer Science, 32:4, IJCS_32_4_13 How Different Choice Strategies Can Affect the Risk Elicitation Process Ari Riabacke, Mona Påhlman, Aron Larsson Abstract This paper
More informationGroup Assignment #1: Concept Explication. For each concept, ask and answer the questions before your literature search.
Group Assignment #1: Concept Explication 1. Preliminary identification of the concept. Identify and name each concept your group is interested in examining. Questions to asked and answered: Is each concept
More informationJohn Quigley, Tim Bedford, Lesley Walls Department of Management Science, University of Strathclyde, Glasgow
Empirical Bayes Estimates of Development Reliability for One Shot Devices John Quigley, Tim Bedford, Lesley Walls Department of Management Science, University of Strathclyde, Glasgow This article describes
More informationCognitive domain: Comprehension Answer location: Elements of Empiricism Question type: MC
Chapter 2 1. Knowledge that is evaluative, value laden, and concerned with prescribing what ought to be is known as knowledge. *a. Normative b. Nonnormative c. Probabilistic d. Nonprobabilistic. 2. Most
More informationAn Escalation Model of Consciousness
Bailey!1 Ben Bailey Current Issues in Cognitive Science Mark Feinstein 2015-12-18 An Escalation Model of Consciousness Introduction The idea of consciousness has plagued humanity since its inception. Humans
More informationUncertainty Characterization: The Role of Hypothesis-Based Weight of Evidence"
Uncertainty Characterization: The Role of Hypothesis-Based Weight of Evidence" Lorenz Rhomberg, PhD Gradient Corporation lrhomberg@gradientcorp.com 617-395-5552 LSRO / CHRA -- 23 May 2007 -- Washington
More informationReporting and commenting on metabolic results. JR Bonham, Sheffield Children s Trust
Reporting and commenting on metabolic results JR Bonham, Sheffield Children s Trust Outline Why is reporting important? What are we trying to achieve? Some features of good practice Some points to bear
More informationBottom-Up Model of Strategy Selection
Bottom-Up Model of Strategy Selection Tomasz Smoleń (tsmolen@apple.phils.uj.edu.pl) Jagiellonian University, al. Mickiewicza 3 31-120 Krakow, Poland Szymon Wichary (swichary@swps.edu.pl) Warsaw School
More informationTHE BAYESIAN APPROACH
WHY BE A BAYESIAN? Michael Goldstein Dept. of Mathematical Sciences, University of Durham,England Abstract This paper presents a discussion of the value of the Bayesian approach for scientific enquiry,
More informationSTIN2103. Knowledge. engineering expert systems. Wan Hussain Wan Ishak. SOC 2079 Ext.: Url:
& Knowledge STIN2103 engineering expert systems Wan Hussain Wan Ishak SOC 2079 Ext.: 4786 Email: hussain@uum.edu.my Url: http://www.wanhussain.com Outline Knowledge Representation Types of knowledge Knowledge
More informationBayesian models of inductive generalization
Bayesian models of inductive generalization Neville E. Sanjana & Joshua B. Tenenbaum Department of Brain and Cognitive Sciences Massachusetts Institute of Technology Cambridge, MA 239 nsanjana, jbt @mit.edu
More informationP(HYP)=h. P(CON HYP)=p REL. P(CON HYP)=q REP
BAYESIAN NETWORKS IN PHILOSOPHY STEPHAN HARTMANN, LUC BOVENS 1. Introduction There is a long philosophical tradition of addressing questions in philosophy of science and epistemology by means of the tools
More informationContext, information s meaning, and perception
8th Conference of the European Society for Fuzzy Logic and Technology (EUSFLAT 2013) Context, information s meaning, and perception Clara Barroso 1 University of La Laguna 2 Associated researcher ECSC
More information