Towards Role Based Hypothesis Evaluation forud Health Data Mining

摘要

Data mining researchers have long been concerned with the application of tools to facilitate andudimprove data analysis on large, complex data sets. The current challenge is to make data miningudand knowledge discovery systems applicable to a wider range of domains, among them health.udEarly work was performed over transactional, retail based data sets, but the attraction of findingudpreviously unknown knowledge from the ever increasing amounts of data collected from the healthuddomain is an emerging area of interest and specialisation. The problem is finding a solution that isudsuitably flexible to allow for generalised application whilst being specific enough to provide functionalityudthat caters for the nuances of each role within the domain. The need for a more granularudapproach to problem solving in other areas of information technology has resulted in the use of roleudbased solutions. This paper discusses the progress to date in developing a role oriented solution toudthe problem of providing for the diverse requirements of health domain data miners and defining theudfoundation for determining what constitutes an interesting discovery in an area as complex asudhealth.