Classification Models for Partially Ordered Sequences

Tripled partially ordered sets

In this paper, we introduce tripled partially ordered sets and monotone functions on tripled partiallyordered sets. Some basic properties on these new dened sets are studied and some examples forclarifying are given.

Sequential Classification on Partially Ordered Sets

A general theorem on asymptotically optimal sequential selection of experiments is presented and applied to a Bayesian classification problem when the parameter space is a finite partially ordered set. The main results include establishing conditions under which the posterior probability of the true state converges to one almost surely, and determining optimal rates of convergence. Properties o...

Ordinal Classification Using Partially Ordered Feature Sets

This paper presents a method to solve an ordinal classification problem where the objects are described with a set of partially ordered features. To construct an ordinal classification model we aggregate partial orders of features by weighting the incidence matrices of partial order graphs. To describe set of possible solutions we use partial order cones corresponding to the incidence matrices ...

An Optimal Strategy for Sequential Classification on Partially Ordered Sets

A decision-theoretic framework is described for sequential classification when the parameter space is a finite partially ordered set. An example of an optimal strategy is then presented. This example establishes that an asymptotically optimal class of experiment selection rules is not necessarily optimal in the given decision-theoretic setting.

Steganalysis Using Partially Ordered Markov Models