In this paper, we propose a cohesion measure to select a cluster of expert preferences in a group decision-making context. Through preference modeling, membership functions are used to express opinions setting the level of agreement over a specific criterion. Taking into account that it is possible to gather a large number of opinions through social media (e.g., facebook, twitter, linkedin, etc.) it is important to handle them properly. Thus, this proposal uses a shape-similarity approach to cluster similar opinions, represents each cluster by means of an interval-valued fuzzy set and provides a cohesion measure to calculate the level of togetherness among membership functions that are present in a cluster. The cohesion measure allows us to discriminate clusters that are relevant to represent expert preferences. An example that illustrates the application of the cohesion measure for expert preferences has been included.
Operational Program Digital Poland, 2014-2020, Measure 2.3: Digital accessibility and usefulness of public sector information; funds from the European Regional Development Fund and national co-financing from the state budget.
Oct 15, 2021
Jul 15, 2021