DISCOVERY OF MULTIPLE LEVEL SEQUENTIAL PATTERNS IN LARGE DATABASES

Knowledge discovery in databases （KDD） is a new emerging area for the research of artificial intelligence and databases,in which the sequential pattern is an important research topic.In the paper, the scope of discovery of the sequential pattern is extended from single level to multiple level,a top down and progressively deepening method is put forward for discovering the sequential pattern in each different level,and three improved methods are proposed for improving the performance of the algorithm.