A natural language generation approach to support understanding and traceability of multi-dimensional preferential sensitivity analysis in multi-criteria decision making

摘要: Multi-dimensional sensitivity analysis is crucial in multi-criteria decision making.Natural language generation techniques increase understanding and traceability.Explanatory concept for multi-dimensional developed.Concept validation by experts online survey.Results show that enhances notably complex situations. Multi-Criteria Decision Analysis (MCDA) enables makers (DM) analysts (DA) to analyse understand situations a structured formalised way. With the increasing complexity of support systems (DSSs), it becomes challenging both expert novice users interpret model results. Natural (NLG) are used various DSSs cope with this challenge as they reduce cognitive effort achieve However, NLG MCDA have so far mainly been developed deterministic or one-dimensional analyses. In paper, textual explanations preferential developed. The key contribution approach provides detailed implications uncertainties Multi-Attribute Value Theory (MAVT). It generates report assesses influences simultaneous separate variations inter-criteria intra-criteria parameters determined within analysis. We explore added value natural an survey. Our results particularly beneficial difficult interpretational tasks.