Tim Pattison

Misclassification Probability Bounds for Multivariate Gaussian Classes

Classification systems based on linear discriminant analysis are employed in a variety of communications applications, in which the classes are most commonly characterized by known Gaussian PDFs. The performance of these classifiers is analyzed in this paper in terms of the conditional probability of misclassification. Easily computed lower and upper bounds on this error probability are presented and shown to provide corresponding bounds on the number of Monte Carlo trials required to obtain a desired level of accuracy. The error probability bounds yield an exact and easily computed expression for the error probability in the case where there are only two classes and a single hyperplane. In the special case where misclassification into a nominated class is independent of all other misclassifications, successively tighter upper and lower bounds can be computed at the expense of successively higher-order products of the individual misclassification probabilities. Finally, bounds are provided on the number of Monte Carlo trials required to improve, with suitably high confidence level, on the confidence interval formed by the error probability bounds.

Enhancing Layout and Interaction in Formal Concept Analysis

Formal Concept Analysis (FCA) derives a multiple-inheritance class hierarchy from a formal context. The number of classes is bounded above by an exponential function of the number of objects and attributes in the context. To support interactive analysis of large formal contexts, this paper exploits a divide-and-conquer technique which discovers hierarchical structure in amenable formal contexts. That hierarchical structure is used to expedite and enhance both the layout of, and user interaction with, the concept lattice. The principal contribution is the dual use of the discovered hierarchical structure for scalable, interactive FCA.

Interfaces for discourse summarisation: a human factors analysis

Empirical studies assessing the effectiveness of novel document interfaces are becoming more prevalent, however relatively little attention has been paid to how such tools could work with less structured documents featuring multiple contributors. Participants in this study used different interfaces to answer questions requiring the exploration of collaborative discourse. User performance was influenced by an interaction of interface, transcript, and question type. Individual differences also impacted on performance with higher education levels and higher general knowledge scores being associated with better task performance. The results also revealed that unnecessary interface functionality can hinder performance.