Novel Approach to Mine Sequential Frequent Pattern

Sequential pattern mining is an important data Mining technique. Mining sequential rules from the sequence database is an important task with wide application. Its use to find frequently occurring ordered events or sub sequence as pattern from sequence database. Sequence can be called as order list of event. If one item set is completely subset of another item set is called sub sequence. Sequential pattern mining is used in various domains such as medical treatments, natural disasters, customer shopping sequences, DNA sequences and gene structures. The problem is to discover the all sequential pattern who satisfy the user specified constraint, from the given sequence database. There are various Sequential pattern mining algorithm like GSP, SPADE, SPAM, PrefixSpan are mainly used to find the relevant sequential frequent pattern from the sequence. All these sequential pattern mining algorithm are generating large set of frequent sequential pattern which are not time and memory efficient. CMRule, ERMiner, and RulrGrowth algorithms generate sequential rule but the method of generation is complicated, memory consumption is also high and it is not time efficient. So the Proposed novel approach i s generating sequential frequent pattern as well as sequential rule in novel Method and it is more efficient in terms of memory and time. Keyword : Sequential pattern mining, Sequence database, Sequential rule Mining.