General Knowledge Items Research Articles

A Chaos-Based Analysis of Discriminative Power of Fuzzy Models

A fractal/chaos analysis and discriminative power evaluation of knowledge bases are presented. They can be used by knowledge engineers to make knowledge acquisition activities more objective. The fractal analysis gives a numerical parameter fractal dimension. It describes certain relations between general and specific knowledge items for example between on-line measurements and general rules of thumbs. The discriminative power analysis is used to quantify one aspect of a knowledge base “quality.” Both formal tools are applicable regardless of types of knowledge bases. The article describes their fuzzy interpretations. A simple demonstrative example a set of 17 fuzzy statements and a realistic fractal recommendation fuzzy versus neural control algorithms are presented in detail.

The Overconfidence Phenomenon as a Consequence of Informal Experimenter-Guided Selection of Almanac Items

The paper argues for an ecological approach to realism of confidence in general knowledge. It is stressed that choice of answer to almanac items and confidence judgments derive from knowledge structures formed by adaptation to a natural environment. People are regarded as well-calibrated to their natural environments and the overconfidence phenomenon is seen as a consequence of the procedures involved in the creation of "traditional" general knowledge items, rather than as the result of a cognitive bias. An experiment is reported showing that when items are informally selected by human "selectors," instructed to select items that differentiate between more and less knowledgeable subjects, we observe poor calibration and the usual overconfidence phenomenon. However, when the item selection is "debiased" and the objects to be compared in the almanac items are selected randomly from a natural environment the overconfidence phenomenon disappears and we observe good calibration, and good resolution. These results support an ecological approach to realism of confidence and suggest that general procedures for debiasing overconfidence may be unwarranted.

A fuzzy upgrading of integrated vague managerial and engineering knowledge

In reality it is not possible to separate managerial and engineering knowledge without a considerable information loss. Realistic integrated conventional managerial/engineering files are rather vague and very sparse. They are not suitable for direct applications of fuzzy reasoning algorithms. A fuzzy upgrading is a retrospective application of the knowledge of engineering methods and integration of large set of specific and rather isolated detailed information items. Any upgrading is highly subjective and ad hoc in nature. The upgrading objectivity can be increased by a set of upgrading algorithms. Two of them (discriminative power evaluation and fractal analysis) are described. The fractal analysis gives a trade-off between general and specific knowledge items and is presented separately as Appendix. A case study (an upgrading of sugar-cane-plant knowledge base) is presented.

A fractal analysis of symbolic and fuzzy knowledge and its engineering applications

Fractal analysis has been used for the study of chaos in a number of static and dynamic systems. The difficulty in applying fractal analysis to symbolic knowledge bases lies in finding a generic notion of scale. The only precondition for the application of fractal analysis is an ability to distinguish between specific and general knowledge items. An analogy with water flow through a pipe network which is topologically identical to the reasoning path under study is used to solve the problem. The main advantages of this methodology are its information non-intensity, simplicity and ability to absorb commonsense heuristics. Therefore it can be easily put to practical use. Two case studies (symbolic/fault tree, fuzzy/ball bearing) are given.

A Fractal Oriented Upgrading of Reliability and Safety Knowledge of Realistic Chemical Engineering Problems

Reliability and safety knowledge is very sparse, Its realistic and objective application is nearly impossible. Additional knowledge is needed to increase the reasoning power of reliability and safety knowledge bases. The conventional fractal analysis has been used for the study of chaos in physical systems. Its possible role in the evaluation of reliability/safety knowledge bases is studied in this paper. The only precondition for the application of fractal analysis is an ability to distinguish between specific and general knowledge items. This enables us to detect a level of inconsistency between mostly subjective additional knowledge items and the existing knowledge bases. The fractal analysis can characterise the knowledge base as one integrated complex. However, knowledge acquisition requires "local" evaluations as well. Therefore a discriminative analysis is used. A realistic man-machine dialogue (an evaluation of the mean time between failures of control valves) supported by the fractal and/or discriminative analysis is presented

Reliability knowledge and fractal evaluation of chaos

Fractal analysis has been used for the study of chaos in a number of two or three dimensional-physical systems. Iits possible role in the evaluation of reliability knowledge bases is studied in this paper. The difficulty in applying fractal analysis to knowledge bases is in finding a generic notion of scale which may be applicable to both numberical and symbolic data. We present here some notions of scale and of detail in relation to a set of fuzzy conditional statements. The only precondition for the application of fractal analysis is an ability to distinguish between specific and general knowledge items. A realistic case study (control valves) is presented in details.

Field dependence and hindsight bias: Cognitive restructuring and the generation of reasons

Cognitive restructuring differences between field dependents and field independents were investigated in the hindsight paradigm using general knowledge items. Experiment 1 found significantly greater hindsight bias in field dependents than in field independents both for items that had been previously rated (“memory”) and for items that had not been previously rated (“hypothetical”). Experiment 2 examined the role of cognitive restructuring more directly via experimental manipulations of reason generation. Hindsight generation of reasons contradicting the correct answer reduced hindsight bias significantly compared with generation of supporting reasons or no reasons. However, this reduction in hindsight bias was much more marked in field dependents than in field independents, despite the fact that field dependents had greater difficulty generating contradictory reasons. Results were discussed in terms of field dependence-independence differences in the spontaneous generation of reasons.

Knowledge, calibration, and resolution: A linear model

Calibration expresses the correspondence between subjective and objective probability, i.e., relative frequency. A subject is perfectly calibrated if for all propositions assigned the probability x ̇ x, x ̇ x of them are true. When proportion correct is plotted against categorized assessments of subjective probability one gets a calibration curve. Resolution expresses the degree to which the subject can sort correct and incorrect items into different categories. A model which approximates the calibration curve by a linear function c( x t ) = a + bx t is suggested, where c( x t ) is the predicted proportion correct when the probability assessment is x t . Perfect calibration means that a = 0, b = 1. When c( x t ) is combined with the distribution of assessments a linear relation •c = a + b x ̄ is obtained; • c is the predicted mean proportion correct and x ̄ is the mean of the probability assessments. The model was tested on (1) individual data from an experiment with 250 general knowledge items and (2) group data from a study by Lichtenstein and Fischhoff (1977, Organizational Behavior and Human Decision Processes, 20, 159–183). The fit between predicted (• c) and observed proportion correct ( c ̄ ) was satisfactory for 31 of 33 subjects in the individual study and for all 19 conditions in the group study. The two studies were compared in some detail with respect to relationships between knowledge on the one hand and calibration and resolution on the other.

Components of probability judgment accuracy: Individual consistency and effects of subject matter and assessment method

An experiment is reported in which subjects assigned probabilities to the outcomes of basketball games and to the truth of general-knowledge items. Three different methods were used for eliciting subjects' probability judgments. Subjects were more successful in selecting answers to the general-knowledge questions than they were in picking basketball game winners. The overall accuracy of their probability judgments for general-knowledge items was superior, too. On the other hand, subjects' judgments about general-knowledge questions were more overconfident, more poorly calibrated, and included greater scatter. One method of probability assessment gave subjects an irrelevant cue. This was found to increase confidence and overconfidence and to hurt calibration. Correlations between measures of performance on general-knowledge questions and basketball predictions showed substantial individual consistency in confidence, but only weak consistency in other components of judgment quality. Theoretical and practical implications are discussed.