Gradual Complex Numbers , local ordering and local aggregations.
Gradual number, Gradual Complex Number, decision making, sectional aggregation, sectional order, aggregation.
Recently, a method based on the Graduated Complex Numbers (NCG) was proposed by the author in order to perform a multicriteria qualitative evaluation of the classification algorithms in a more flexible and subjective way. This new approach differs from existing measures because it attributes linguistic terms to the performance of each classifier and associates the same position in the classification ranking to algorithms with very close performances. However, because we do not have a total order on these numbers, the method uses a calculation where a lot of information about the NCG is lost. In addition,
the function used to "aggregate"information from NCGs does not have the characteristics of an aggregation function, which can lead to problems in the final result. In order to overcome these problems, the objective of this project is to introduce the notion of local orders and aggregations that will be applied to NCGs.