Abstract-In non-stationary learning, we require a predictive model to learn over time, adapting to changes in the concept if necessary. A major concern in any algorithm for non-stationary learning is its rate of adaptation to new concepts. When tackling such problems with ensembles, the concept of diversity appears to be of significance. In this paper, we discuss how we expect diversity to impact the rate of adaptation in non-stationary ensemble learning. We then analyse the relation between voting margins and a popular measure of diversity, KW variance, and use the similarities between them to draw some useful conclusions regarding ensemble adaptivity.

Abstract. Scientific workflows may include automated decision steps, for instance to accept/reject certain data products during the course of an in silico experiment, based on an assessment of their quality. The trustworthiness of these workflows can be enhanced by providing the users with a trace and explanation of the outcome of these decisions. In this paper we present a provenance model that is designed specifically to support this task. The model applies to a particular type of subworkflow that is compiled automatically from a high-level specification of user-defined, quality-based data acceptance criteria. The keys to the effectiveness of the approach are that (i) these sub-workflows follow a predictable pattern structure, (ii) the purpose of their component services is defined using an ontology of Information Quality concepts, and (iii) the conceptual model for provenance is consistent with the ontology structure.