International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems

In order to represent a "human-like" behavior-decision ability in an autonomous mobile robot it is necessary to describe behavioral sequences while taking into account the recognition of ambiguous states. Fuzzy algorithms capable of expressing sequence flow including a mixture of crisp and fuzzy processing provide an efficient behavior-decision algorithm. We have already proposed the Modified Fuzzy Algorithm (MFA) with tuning functions where branch is selected according to a controlled threshold. Results from computer simulations were also shown to prove the effectiveness of this method for giving intelligent mobile robots the autonomous macro behavior decision abilities like humans. In this paper we extend our previous work by using Fuzzy Petri Nets (FPN) to express a fuzzy algorithm. Ambiguous state transition is evaluated by marking change of the fuzzy truth tokens. The state analysis mechanism supplied by the Petri net makes possible the objective and global evaluation of the ambiguity in multi-firing of sequential rules. This new method assures that we can express the fuzzy algorithm in which explosions of fuzziness do not occur. For illustration purposes we present simulation results obtained using this method for the behavior-decision fuzzy algorithm for an autonomous mobile robot.

In this paper, a careful analysis of interval-valued possibilistic knowledge bases indicates that there exists a natural probability distribution over the set of orderings of formulae compatible with the weights given in the knowledge base. We propose a new view, by which a possibilistic knowledge base can be considered in term of such probability distribution. It reveals some interconnections between probabilistic and possibilistic logics. We show that the principle of minimum specificity, widely used in possibilistic logic, is a special case of the principle of maximum likehood (at least, from the standpoint of nonmonotonic reasoning). We propose a formula to calculate probability for a defeasible conclusion. Moreover, the proposed view seems to be useful for other practical purposes. As an example, we apply it to a traditional problem of fusion of possibilistic knowledge from many sources and derive a new solution.

The paper shows the existence of a canonical separated hierarchical decomposition of the Choquet integral over a finite set. The decomposed system is a hierarchical combination of Choquet integrals with mutually disjoint domains, and equivalent to the original Choquet integral. The paper also gives canonical overlapped hierarchical decompositions of the Choquet integral over a semiatom.

In this paper, an Integrated C-Object Tool, namely ICOT, is proposed for knowledge-based programming. A major drawback of current rule-based expert system languages is that they have difficulty in handling composite objects as a unit of inference. An object-oriented model is a powerful alternative to complement the drawback. Each of these alone cannot capture all the semantics of knowledge, particularly in complex engineering domains. For a knowledge-based approach to be effective, both the object-oriented paradigm and the rule-based mechanism may need to be integrated into one framework. The framework may also need to support manipulation of fuzzy knowledge to model the real world as close as possible. Three types of fuzzy information are identified, and a proper way of representing and inferencing them is developed. ICOT provides a new framework into which rule-based deduction, object-oriented modeling, and fuzzy inferencing are combined altogether. This can become especially useful for developing knowledge-based engineering applications.

We find a fifth approximation of the Just Intonation which generalizes Equal Temperament. The intervals causing a dilemma are the second and the minor seventh and the tritone because they are unambiguous in Just Intonation (the relative frequencies 10/9, 9/8, 8/7 and 7/4, 16/9, 18/10 and 45/32, 64/45, respectively). If we do not consider the second and seventh with the relative frequencies 8/7 and 7/4, respectively, all the music intervals in this approximation either coincide with he Just Intonation interval values (the octave, fifth, fourth, second (9/8) and the minor seventh (16/9)) or are exactly the one comma distant from the corresponding Just Intonation intervals. This comma is 32 805/32 768 ≈ 1.00112915, which is less than the ratio of frequencies of the perfect and the equal tempered fifths (≈ 1.00112989).

In induction graphs methods such as C4.5¹ or SIPINA², taking continuous attributes into account needs particular discretization procedures. In this paper, we propose on the one hand, an axiomatic leading to a set of criteria which can be used for continuous attributes discretization, and on the other hand, a method of discretization called FUSINTER. The results obtained by FUSINTER are compared to those obtained by techniques developed by Fayyad and Irani³ and Kerber⁴ and they have proved better for the majority of the examples studied.

This section is maintained by Vladik Kreinovich. Please send your abstract (or copies of papers that you want to see reviewed here), to vladik@cs.utep.edu, or by regular mail to: Vladik Kreinovich, Department of Computer Science, University of Texas at El Paso, El Paso, TX 79968, USA.