Ranking association rules by clustering through interestingness

• Autor(a) principal: de Carvalho, Veronica Oliveira [UNESP]
• Data de Publicação: 2018
• Outros Autores: de Paula, Davi Duarte [UNESP], Pacheco, Mateus Violante [UNESP], dos Santos, Waldeilson Eder [UNESP], de Padua, Renan, Rezende, Solange Oliveira
• Tipo de documento: Artigo de conferência
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.1007/978-3-030-02837-4_28; http://hdl.handle.net/11449/187266
• topic: Association rules; Clustering; Objective measures; Post-processing

The association rules (ARs) post-processing step is challenging, since many patterns are extracted and only a few of them are useful to the user. One of the most traditional approaches to find rules that are of interestingness is the use of objective measures (OMs). Due to their frequent use, many of them exist (over 50). Therefore, when a user decides to apply such strategy he has to decide which one to use. To solve this problem this work proposes a process to cluster ARs based on their interestingness, according to a set of OMs, to obtain an ordered list containing the most relevant patterns. That way, the user does not need to know which OM to use/select nor to handle the output of different OMs lists. Experiments show that the proposed process behaves equal or better than as if the best OM had been used.