Survey of derived high utility pattern mining algorithms
In the field of data mining,the task of high utility pattern mining has a high theoretical research value and a wide range of practical application scenarios.A series of derived high utility patterns are proposed for variable applications.Firstly,the high average utility pattern mining algorithms are discussed from the perspective of key technologies,mainly including Apriori-based,tree-based,list-based,projection-based and data structure-based algorithms.Secondly,high utility pattern mining algorithms containing negative utility based on complete set,concise set and integrated pattern are analytically discussed.Then,the extended high utility pattern algorithms are outlined and summarized from three aspects,including fuzzy high utility patterns,correlated high utility patterns and other emerging high utility patterns.Finally,in view of the shortcomings of the current research direction,the next research directions are given.
derived high utility patternhigh average utility patternnegative utilityfuzzy high utility patterncorrelated high utility patternsurvey
国家科技期刊平台
NSTL
万方数据
