Mining Positive and Negative Sequential Pattern in Incremental Transaction Databases

Abstract

and negative sequential patterns mining is used to discover interesting sequential patterns in a incremental transaction databases, and it is one of the essential data mining tasks widely used in various application fields. Implementation of this approach, construct tree for appended transactions (new upcoming data) and will merge this tree with existing tree (tree of existing transactions) to get the Updated tree. Positive and negative sequential Patterns mining is an aim to find more interesting sequential patterns, considering the minimum support of each data item in a sequence database. Generally, the generation order of data elements is considered to find sequential patterns. Positive sequential patterns states that these items were occur with one and another. Actually the absence of certain itemset may imply the appearance of other itemsets as well. The absence of itemsets thus is becoming measurable in many applications. Negative sequential patterns could assist product recommendation systems to make more accurate decisions. This approach will reduce the mining time for incremental database if the Existing database has lots of transactions and Appended database having few transactions.