An efficient algorithm for mining high utility patterns from incremental databases with one database scan

摘要: High utility pattern mining has been actively researched as one of the significant topics in data field since this approach can solve limitation traditional that cannot fully consider characteristics real world databases. Moreover, database volumes have bigger gradually various applications such sales retail markets and connection information web services, general methods for static databases are not suitable processing dynamic extracting useful from them. Although incremental approaches suggested, previous need at least two scans irrespective using any structure. However, with multiple actually adequate stream environments. In paper, we propose an efficient algorithm high patterns scan based on a list-based structure without candidate generation. Experimental results synthetic datasets show proposed outperforms phase construction