PERCEPTIONS
Mathematical structures based on unbalanced and hesitant linguistic descriptions to improve human-machine interaction
- Principal investigators: Núria Agell and Mònica Casabayó
- Research group: Research Group on Judgements & Decisions in Marketplace (JUICE)
- Funding body: MICINN-MCIU
About the project
Hesitant linguistic assessments involving unbalanced and multi-granular information, are considered appropriate to grasp the uncertainty and imprecision which exists in human reasoning when expressing opinions and preferences. To better process this type of information for easier analysis and interpretation, this project proposes the development of perceptual maps as new mathematic structures to be applied in multiple-criteria systems and social networks frameworks. This will allow us to improve human-machine interaction and revert in a better design of decision-aiding systems and social interaction platforms. In addition, the project will be focusing its attention to design solutions for advanced managerial problems involving human-machine interaction.
First, the project will analyze new perceptual-based maps able to capture differences between unbalanced linguistic assessments. Second, new algebraic structures will be defined to deal with multi-perceptual group decision-making contexts where each decision-maker has his own linguistic descriptions to assess each criterion or alternative. Finally, a methodology to aggregate unbalanced linguistic information based on different perceptual maps will be developed. This will allow us to consider new distances in hesitant and unbalanced linguistic terms sets and be able to measure consensus and consistency measures in decision-making processes.
Our two main goals for this project are to make a strong contribution and advancement in these theoretical approaches in the area of cognitive computation and decision-making, and, in parallel, our second goal is to present real applications in social and business contexts in which the methodological advances presented could be useful. Both objectives will be considered in situations framed in multi-criteria and group-decision systems and situations in which the information comes from users participating in social networks or platforms.
