关键词: 稀有序列模式挖掘, 自适应间隙, 一次性条件, 支持度计算, 剪枝策略

Abstract: Rare sequential pattern mining aims to discover infrequent and important patterns in sequence databases. However, current sequential pattern mining methods mostly determines whether a pattern occurs in a sequence or not, ignoring the repetition of the pattern in the sequence, that is, the user's level of interest, resulting in bias in the mining results. To tackle this issue, a rare sequential pattern mining method with adaptive gap under one-off condition was proposed, which calculates the number of repetitions of the pattern in the sequence using one-off condition and reflects the sequence features using adaptive gaps. To avoid the inefficient sequential traversal of the original database in the support calculation process of traditional algorithms, an inverted index structure was established, which stores each item and its location in the original database, thereby avoiding the problem of redundant sequential traversal of the original database and improving the efficiency of support calculation. In the process of candidate pattern generation, the pattern join strategy was used to generate candidate patterns. To further reduce the number of candidate patterns, a pruning strategy was proposed, thereby improving the mining speed. Experiments were conducted on five real datasets. The experimental results showed that the proposed algorithm outperforms other competitive algorithms, thus verify the superiority of the proposed algorithm.

Key words: rare sequential pattern mining, adaptive gap, one-off condition, support calculation, pruning strategy