An approach to optimize Fuzzy Time-Interval Sequential Patterns using Multi-Objective Genetic Algorithm

Sequential pattern mining, which discovers frequent subsequences as patterns in a sequence database, is an important data-mining problem with broad applications. From these discovered sequential patterns, we can discover the order of the patterns; however, they cannot tell us the time intervals between successive patterns. Accordingly, Chen et al. have proposed a fuzzy timeinterval (FTI) sequential pattern mining algorithms, which reveals the time intervals between successive patterns [12][13]. In this paper, we contributed to the ongoing research on FTI sequential pattern mining by proposing a multi objective Genetic Algorithm (GA) based method. Fuzzy solves the sharp boundary problem and the refinement ability of GA helps to find the global optimum FTI sequential patterns. Our approach uses two measures as objectives, namely: Confidence and Coverage to prune the traditional Apriori algorithm. The main objective is to achieve maximum confidence and maximum coverage in the FTI sequential patterns. The paper defines the confidence of the FTI sequences, which is not yet defined in the previous researches. The main advantage of the proposed algorithm is the use of fuzzy genetic approach to discover optimized sequences in the network traffic data to classify and detect intrusion.