Use of data analytics to build intuitive decision models – an approach to indirect derivation of criteria weights based on discordance related preferential information

Data on past and current decisions can be utilised to enhance the decision-making process by automating or making problem solving more intuitive. is either extracted from distributed sources repositories, obtained with regression analysis holistically assessed alternatives human judgements. One of possible advanced approaches encourage intuitive aims at inferring criteria weights decision model regard correlations between preferential parameters, in such a way that objective inner information consolidated personal knowledge experience. This relevant because task specifying cognitively demanding represents key aspect each model. The paper first discusses notation infrastructure exchange models handle underlying mechanisms indirect weight derivation. As main contribution research, method for inference veto-related proposed, which selective strengths veto degrees are calculated compare magnitudes veto, while assessments used determine influence deterioration rankings categories into sorted, respectively. Strengths non-compensatory then projected compensatory weights. experimental study reveals characteristics indirectly derived veto. Several quality factors considered, as validity weights, accuracy results, richness ability discriminate conflicting alternatives. Weights also compared standard ROC RS surrogate approach generalised both common problematics ranking sorting.