A Stochastic Approach for Finding Optimal Context in a Contextual Pattern Analysis Task

This article concerns contextual pattern analysis tasks. As different contexts give different performances, models for finding the optimal context are revisited here. Random field models for the input data are assumed. An underlying random field is represented by a set of parameters capturing the spatial dependence. Next, a Bayesian approach is revisited to develop a decision rule for choosing appropriate context. The relevance of this approach is explored for three pattern analysis tasks, namely, handwriting analysis, image compression, and word sense disambiguation.