A Selection of Useful Patterns Based on Multi-Criteria Analysis Approach
Fatima Zahra El Mazouri, Mohammed Chaouki Abounaima, Khalid Zenkouar · 2017
Pattern discovery is one of the most important tasks in data mining, many works are developed in this context where we can notice the problem of 'pattern explosion' which make taking a decision about useful pattern more difficult. the goal of our study is to make an improvement in the process of extracting useful patterns from data, called also 'pattern mining'. The aim of this article is to propose an approach to select useful patterns from a set of patterns by using multicriteria approach. To do this, we will use the famous multicriteria analysis method called ELECTRE, particularly ELECTRE I, in order to have a selection of the most relevant patterns according to proposed criteria.