基于WIdT-Tree的带权频繁模式挖掘算法.pdfVIP

M ay COM PU T ER ENGINEERING A ND DESIGN Vol No WIdTTree the difference of tw o T idsets dT idset dT idset WIT T ree WIdT T ree dT idset WIdT T ree T P A doi issn j A lgorithm f or mining w eighted f requent patterns based on W IdT T ree T A N Long QIN Qibing College of Computer Science and T echnology Heilongjiang U niversity Harbin China Abstract T o improve the time and space efficiency in the process of w eighted f requent pattern mining the concept of dT idset the difference of tw o T idsets w as defined On the basis of dT idset WIT T ree storage structure w as improved and optimized and an efficient algorithm based on WIdT T ree w as proposed to discover w eighted f requent patterns Besides the divide and conquer strategy and the theorems of dT idset w ere used to calculate the w eighted support and computational efficiency w as greatly improved T he w eighted f requent patterns w ere stored on WIdT T ree to reduce the space consumption Algorithms w ere tested and compared on real and synthetic datasets Ex perimental results show that the proposed algorithm has excellent time and space efficiency and better scalability Key words w eighted f requent patterns dT idset divide and conquer strategy w eighted support T ao F WA R 0 association rule mining ARM dow nw ard closure property DCP DCP Peng supportconfidence