Pure Incremental Approach for Sequential Pattern Mining

In data mining, mining sequential pattern from a very huge amount of database is very useful in many applications. Most of sequential pattern mining algorithms works on static data means the database should not change. But the databases in today’s real world application do not have static data, rather they are incremental databases. New transactions are added at some intervals of time in database. For updated database, the algorithm actually needs to be executed again for whole sequence database. So those approaches are not appropriate to use, for that the algorithm with incremental approach should be modelled and used. In this paper analysis of existing approaches for finding sequential pattern mining, and the survey is helpful in forming a new model or improving some existing approach to handle incremented database & obtain sequential patterns out of them. In this a proposed a model that is totally incremental approach, which we call pure incremental approach. This proposed pure incremental mining is used for mining the frequent sequences for sequence database.