Displaying 21 – 40 of 384

Showing per page

Text document classification based on mixture models

Jana Novovičová, Antonín Malík (2004)

Kybernetika

Finite mixture modelling of class-conditional distributions is a standard method in a statistical pattern recognition. This paper, using bag-of-words vector document representation, explores the use of the mixture of multinomial distributions as a model for class-conditional distribution for multiclass text document classification task. Experimental comparison of the proposed model and the standard Bernoulli and multinomial models as well as the model based on mixture of multivariate Bernoulli distributions...

The application of generalised constraints to object-oriented database models.

Guy de Tré, Rita de Caluwe (2000)

Mathware and Soft Computing

A formal framework for a generalised object-oriented database model is presented, which is able to cope with fuzzy and uncertain information. This model is obtained as a generalisation of a crisp object-oriented database model, which is consistent with the ODMG de facto standard and is built upon an algebraic type system and a constraint system. Generalised constraints have been used to enforce integrity rules and to specify the formal semantics of the database model.

The Bhattacharyya metric as an absolute similarity measure for frequency coded data

Frank J. Aherne, Neil A. Thacker, Peter I Rockett (1998)

Kybernetika

This paper highlights advantageous properties of the Bhattacharyya metric over the chi-squared statistic for comparing frequency distributed data. The original interpretation of the Bhattacharyya metric as a geometric similarity measure is reviewed and it is pointed out that this derivation is independent of the use of the Bhattacharyya measure as an upper bound on the probability of misclassification in a two-class problem. The affinity between the Bhattacharyya and Matusita measures is described...

The box parameter for words and permutations

Helmut Prodinger (2014)

Open Mathematics

The box parameter for words counts how often two letters w j and w k define a “box” such that all the letters w j+1; ..., w k−1 fall into that box. It is related to the visibility parameter and other parameters on words. Three models are considered: Words over a finite alphabet, permutations, and words with letters following a geometric distribution. A typical result is: The average box parameter for words over an M letter alphabet is asymptotically given by 2n − 2n H M/M, for fixed M and n → ∞.

The branch and bound algorithm for a backup virtual path assignment in survivable ATM networks

Krzysztof Walkowiak (2002)

International Journal of Applied Mathematics and Computer Science

Issues of network survivability are important, since users of computer networks should be provided with some guarantees of data delivery. A large amount of data may be lost in high-speed Asynchronous Transfer Mode (ATM) due to a network failure and cause significant economic loses. This paper addresses problems of network survivability. The characteristics of virtual paths and their influence on network restoration are examined. A new problem of Backup Virtual Path Routing is presented for the local-destination...

Currently displaying 21 – 40 of 384