Attitude-based entropy function and applications in decision-making. (September 2021)

• Record Type: Journal Article
• Authors: Aggarwal, Manish
• Is Part Of: Engineering applications of artificial intelligence. Volume 104(2021)
• Publication Date: 2021-09
• Subjects: Entropy -- Attitude -- Uncertainty -- Decision making; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2021.104290
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: The popular entropy functions are rigorously analysed in the context of uncertainty in the real world decision making. Based on the findings, a new entropy function is introduced specifically for the human decision making. The proposed function considers an agent's degree of sensitivity towards uncertainty, i.e., the tendency to exaggerate or downplay the inherent uncertainty. The proposed entropy function is equipped to deal with both the subjective and probabilistic uncertainties alike, which are often interlinked in a decision-making context. The properties of the proposed entropy function are rigorously studied. A real case-study in portfolio diversification highlights the usefulness of the entropy function. It was found that the attitude plays a profound role, when there are a large number of uncertain systems (portfolios) to compare and choose from, or when the portfolios are more diversified.