Mining Distributed Databases for Negative Associations from Regular and Frequent Patterns

Abstract

Most of the transactional data stored in distributed databases, especially when region wise business conducted. Extensive knowledge about the business is hidden in these databases, the discovery of which will give immense power to the decisionmakers. Most of the decisions are taken around frequently occurring patterns and the positive associations that exist among those patterns. Mining a single database can yield the patterns of sales happening at different distributed locations. The regularity of the occurrence of sales pattern is equally as important as frequent patterns. Negative associations also reveal interesting discovery about the business in addition to positive associations and yet time negative associations among the regularly and frequently occurring patterns having a heavy bearing on the Business. It is also important to study the occurrence of the patterns at all the distributed locations and determine the regularity or global nature of the patterns.In this paper, Algorithms presented using which mining of negative association that exist among regular and frequent patterns mined presented considering both local and Global Locations.