Second-Order Belief Hidden Markov Models

TL;DR

This paper introduces a second-order Hidden Markov Model utilizing belief functions, extending previous belief HMMs from first-order to capture more complex temporal dependencies in pattern recognition.

Contribution

It presents the novel development of second-order belief HMMs, expanding the applicability of belief function-based models beyond first-order systems.

Findings

First implementation of second-order belief HMMs

Enhanced modeling of temporal dependencies

Potential for improved pattern recognition accuracy

Abstract

Hidden Markov Models (HMMs) are learning methods for pattern recognition. The probabilistic HMMs have been one of the most used techniques based on the Bayesian model. First-order probabilistic HMMs were adapted to the theory of belief functions such that Bayesian probabilities were replaced with mass functions. In this paper, we present a second-order Hidden Markov Model using belief functions. Previous works in belief HMMs have been focused on the first-order HMMs. We extend them to the second-order model.