In the framework of models generated by compositional expressions, we solve two topical marginalization problems (namely, the single-marginal problem and the marginal-representation problem) that were solved only for the special class of the so-called “canonical expressions”. We also show that the two problems can be solved “from scratch” with preliminary symbolic computation.
Efficient computational algorithms are what made graphical Markov models so popular and successful. Similar algorithms can also be developed for computation with compositional models, which form an alternative to graphical Markov models. In this paper we present a theoretical basis as well as a scheme of an algorithm enabling computation of marginals for multidimensional distributions represented in the form of compositional models.
This paper studies the problem of marginalizing convex polytopes of probabilities represented by a set of constraints. This marginalization is obtained as a special case of projection on a specific subspace. An algorithm that projects a convex polytope on any subspace has been built and the expression of the subspace, where the projection must be made for obtaining the marginalization, has been calculated.
Firstly we present a geometric interpretation of interval-valued fuzzy sets. Secondly, we apply the method of least squares to the fuzzy inference rules when working with these sets. We begin approximating the lower and upper extremes of the membership intervals to axb type functions by means of the method of least squares. Then we analyze a technique for evaluating the conclusion of the generalized modus ponens and we verify the fulfillment of Fukami and alumni axioms [9].
Based on rough set theory many algorithms for rules extraction from data have been proposed. Decision rules can be obtained directly from a database. Some condition values may be unnecessary in a decision rule produced directly from the database. Such values can then be eliminated to create a more comprehensible (minimal) rule. Most of the algorithms that have been proposed to calculate minimal rules are based on rough set theory or machine learning. In our approach, in a post-processing stage,...
Information retrieval in information systems (IS) with large amounts of data is not only a matter of an effective IS architecture and design and technical parameters of computer technology used for operation of the IS, but also of an easy and intuitive orientation in a number of offers and information provided by the IS. Such retrievals in IS are, however, frequently carried out with indeterminate information, which requires other models of orientation in the environment of the IS.