Fault Tolerance Patterns Mining in Dynamic Databases

Mining of frequent patterns in database has been studied for several years. However, real-world data tends to be dirty and frequent pattern mining which extracts patterns that are absolutely matched is not enough. An approach, called frequent fault-tolerant pattern (FT-pattern) mining, is more suitable for extracting interesting information from real-world data that may be polluted by noise. Previous research on frequent fault-tolerant pattern mining has been widely studied. However, all of the researches focus on static database. In this paper, we propose an efficient framework to analyze the frequent FT-patterns mining in dynamic database. To avoid re-scanning the whole database, beside of keeping the fault-tolerance pattern, we will also keep the potential fault-tolerance pattern that has higher possibility of becoming a fault-tolerance pattern. The experimental results show that by re-using the existing pattern that had been generated, the proposed algorithms are highly efficient in terms of execution time and maximum memory usage for mining fault-tolerance frequent pattern in dynamic database compare to FFM algorithm.