.png){kind=icon}
.png){kind=icon}
Return
Dynamic-R: a challenge-free method for rating problem statements
.png){kind=icon}
PRE
AI
DOI：10.1007/s10479-023-05276-8.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this paper, we are interested in decision aiding problems, aiming at rating a set of objects with respect to several dimensions, called criteria. A rating problem statement consists on partitioning a set of objects into predefined ordered equivalence classes, called categories, identified by ratings. Rating problems are widely studied in the literature, either based on the utility theory, rough sets or the majority principle. The existing methods based on the majority principle present some disadvantages potentially leading to an unconvincing rating because challenged by contradictory pairwise comparisons. In this work, we present a new method providing a convincing (challenge-free) rating over a set of studied objects, based on the aggregation of positive and negative reasons, respectively supporting and opposing a rating. The method exploits comparisons among the objects and the profiles characterizing the categories as well as comparisons among the objects.
Keywords:
Multiple criteria decision analysis
Rating problem statements
Decision support systems
Algorithmic decision theory
Journal
IF:
4.5
Papers:
8.0K
Citations:
2.1W