Maximal Sequential Pattern Mining Based on Simultaneous Monotone and Anti-monotone Constraints

The main challenge of mining sequential patterns is the high processing cost of support counting for large amount of candidate patterns, and a lot of patterns are not interesting to users. In this paper, a novel algorithm MSMA (maximal sequential pattern mining based on simultaneous monotone and anti-monotone constraints) incorporating both maximal and constraint-based sequential pattern mining in mining process is proposed. It allows the efficient mining of sequential patterns when both monotone and anti-monotone constraints are simultaneously pushed in mining process at different strategic stages. Our experiment shows that MSMA is an efficient algorithm for handling simultaneous monotone and anti-monotone constraints.