Pattern recognition is seen as a major challenge within the field of data mining and knowledge discovery. For theudwork in this paper, we have analyzed a range of widely used algorithms for finding frequent patterns with theudpurpose of discovering how these algorithms can be used to obtain frequent patterns over large transactionaluddatabases. This has been presented in the form of a comparative study of the following algorithms: Aprioriudalgorithm, Frequent Pattern (FP) Growth algorithm, Rapid Association Rule Mining (RARM), ECLAT algorithmudand Associated Sensor Pattern Mining of Data Stream (ASPMS) frequent pattern mining algorithms. This studyudalso focuses on each of the algorithm’s strengths and weaknesses for finding patterns among large item sets inuddatabase systems.
展开▼
