Selecting Relevant Association Rules From Imperfect Data - Archive ouverte HAL Access content directly
Conference Papers Year : 2019

Selecting Relevant Association Rules From Imperfect Data


Association Rule Mining (ARM) in the context of imperfect data, e.g., imprecise data, missing data, has received little attention so far despite the prevalence of such data in a wide range of real-world applications. In this work, we present a ARM approach that can be used to handle imprecise data and derive imprecise rules. Based on the belief functions framework and Multiple Criteria Decision Analysis, the proposed approach relies on a selection procedure for identifying the most relevant rules while considering information characterizing their interest-ingness. The several measures of interestingness defined for comparing the rules as well as the selection procedure, are presented. We also show how a priori knowledge about attribute values defined into domain tax-onomies can be used to reduce the search space and the complexity of the mining process, in addition to help identifying relevant rules for the domain of interest. Our appoach is illustrated using a concrete simplified case study related to humanitarian projects analysis.
Fichier principal
Vignette du fichier
SUM2019_vf.pdf (316.78 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02309641 , version 1 (24-02-2020)



Cécile L 'Héritier, Sébastien Harispe, Abdelhak Imoussaten, Gilles Dusserre, Benoit Roig. Selecting Relevant Association Rules From Imperfect Data. 13th international conference on Scalable Uncertainty Management (SUM 2019), Dec 2019, Compiègne, France. ⟨10.1007/978-3-030-35514-2_9⟩. ⟨hal-02309641⟩
142 View
125 Download



Gmail Facebook X LinkedIn More