A Fast Algorithm for Mining Sequential Patterns from Large Databases

Mining sequential patterns from large databases has been recognized bymany researchers as an attractive task of data mining and knowledgediscovery. Previous algorithms scan the databases for many times, whichis often unendurable due to the very large amount of databases. In thispaper, the authors introduce an effective algorithm for mining sequentialpatterns from large databases. In the algorithm, the original databaseis not used at all for counting the support of sequences after the firstpass. Rather, a tidlist structure generated in the previous pass isemployed for the purpose based on set intersection operations, avoidingthe multiple scans of the databases.