Sie befinden Sich nicht im Netzwerk der Universität Paderborn. Der Zugriff auf elektronische Ressourcen ist gegebenenfalls nur via VPN oder Shibboleth (DFN-AAI) möglich. mehr Informationen...
Discovering High Utility Itemsets Using Set-Based Particle Swarm Optimization
Ist Teil von
Advanced Data Mining and Applications, p.38-53
Ort / Verlag
Cham: Springer International Publishing
Link zum Volltext
Quelle
Alma/SFX Local Collection
Beschreibungen/Notizen
Mining high utility itemsets (HUIs) is a hot research topic in data mining. Algorithms based on evolutionary computation are attracting increasing attention because they have the advantage of avoiding the combinatorial explosion of the HUI search space. Among evolutionary methods used for mining HUIs, particle swarm optimization (PSO) is the most popular. Existing PSO-based HUI mining (HUIM) algorithms transform positions according to the result of applying the sigmoid function to the velocity. In this paper, we propose an HUIM algorithm based on set-based PSO (S-PSO) called HUIM-SPSO, which mainly considers elements in positions whose velocities are high. We introduce the modeling of HUIM using S-PSO, and explain HUIM-SPSO in detail. To reflect the diversity of the mining results, we propose the measure of the bit edit distance. Extensive experimental results show that the HUIM-SPSO algorithm is efficient and can discover more HUIs with a high degree of diversity.