Multidimensional multigranularity data mining for health-care service management

Abstract

Data Mining is getting increasingly important for discovering association patterns for health service innovation and Customer Relationship Management (CRM). Yet, there are deficits of existing data mining techniques. First of all, most of them perform a plain mining based on a predefined schemata through the data warehouse; however, a re-scan must be done whenever new attributes appear. Second, an association rule may be true on a certain granularity but fail on a smaller one and vice versa. Last but not least, they are usually designed to find either frequent or infrequent rules. In this article, we are going to invent more efficient and accurate approach with novel data structure and multidimensional mining algorithm to explore association patterns on different granularities and to find out portfolios of health-care service management. © 2015 Taylor & Francis Group, London.