.png){kind=icon}
.png){kind=icon}
Return
Mining closed high utility patterns with negative utility in dynamic databases
.png){kind=icon}
PRE
AI
DOI：10.1007/s10489-022-03876-8.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
High utility itemsets mining algorithms with negative utility have more practical applications because they can handle datasets containing negative items. Existing algorithms that consider negative items assume that the database is static and contain a lot of redundant itemsets information in the result set. To solve these problems, a first algorithm for mining closed high utility itemsets containing negative items in dynamic databases, called CHUInd, is proposed. Dynamic list index structure designed in the algorithm to quickly access and update the information stored in the list based on the index values. Memory reuse strategy is applied to reduce memory usage and quickly update item information during batch insertion. Extensive experimental evaluations on real datasets show the efficiency as well as the feasibility of the algorithm, which exhibits excellent performance in terms of both runtime memory usage.
Keywords:
High utility pattern mining
Negative utility
Index
Utility list
Closed itemsets