Constraint based Sequential Pattern Mining in Time Series Databases - A Two Way Approach

Abstract

Most of the patter search algorithms in literature are mostly string based and do not concentrate on finding sequential patterns with constraints in a given database. Also in query languages such as SQL or MySQL, the select clause does not allow the use of the non-aggregate functions as part of query compilation. The objective of this work is to propose a pattern mining algorithm which may be embedded into SQL or MySQL Query languages so that we can search for presence of sequential pattern in the database. The algorithm makes use of sliding sequential patterns with three look-ahead elements in the event of any mismatch.