A Practical Multi-Ontology Approach to Knowledge Exploration
Gary H. Merrill
Abstract: Babylon Knowledge Explorer is an open source based platform for developing knowledge exploration and data mining applications. As a strongly ontology- driven information system, it rests on an approach and solution to the challenges of using pre-existing large ontologies in scientific domains. The aim of this paper is to describe the model of ontology representation employed by BKE and how this model supports a strategy and methodology facilitating application of these pre-existing ontologies to real-world problems of knowledge discovery in large document corpora and databases. Suficient detail of design and architecture is provided, together with some sketch of the open source implementation, to allow others to make use of this approach.
Adaptive Model-Based Designs for Dose-Finding Studies
Vladimir Dragalin and Valerii Fedorov
Abstract: We propose a new adaptive procedure for dose-finding in clinical trials when both efficacy and toxicity responses are available. We model the distribution of this bivariate binary endpoint using either Gumbel bivariate logistic regression or Cox bivariate binary model. In both cases, the analytic formulae for the Fisher information matrix are obtained, that form the basis for derivation of the locally optimal and adaptive designs.
Optimal Design for Beta Distributed Responses
Yuehui Wu, Valerii V. Fedorov, and Kathleen J. Propert
Abstract: Whenever a response is naturally con¯ned to a ¯nite set (such as a visual analog scale for pain severity) the Beta distribution provides a simple and °exible probability distribution to model such a response. The parameters of the distribution can then be related to covariates such as dose in a clinical trial through the generation of a Beta regression model. We explore locally optimal designs for this class of regression models focusing mainly on minimization of the generalized variance of maximum likelihood estimators (D-optimality). Optimal designs and sensitivity to misspeci¯cation of model parameters are examined using a candidate points searching algorithm. Although formally the model assumes that the response is continuous, it provides a parsimonious approximation for ordinal data when there is a relatively large number of categories. The resulting estimators and optimal designs are simpler and may offer more ease in interpretation than those derived from models for ordered categorical outcomes. The proposed methods are applied to data from a clinical trial.
