A perceptual computer for hierarchical portfolio selection based on interval type-2 fuzzy sets

Today’s advancements have made financial markets accessible to everyone; hence, portfolio selection has become an individualized decision-making problem without the need of being highly educated. Individual judgments, however, are subjective and are influenced by the individual’s background, experience, and views. Existing methods do not account for the personalized criteria and preferences or do not let people express their preferences and assessments using words or terms from natural languages. This paper proposes a framework for an individualized hierarchical portfolio selection system based on perceptual computing. The proposed method assists individuals to rank and select portfolios based on their personalized criteria and preferences and according to their subjective assessments. In this paper, words that are used to express one’s preferences, evaluations, and weights are modeled with interval type-2 fuzzy sets (IT2FS), which allows handling different levels of linguistic uncertainties with manageable computational complexities. The proposed method is applicable to any set of criteria and sub-criteria devised to evaluate portfolios. Moreover, it enables different individuals with different expertise levels to evaluate those criteria. The conducted experiments show that the proposed method, compared to other methods, is reliable and robust to the linguistic uncertainties and provide plausible recommendations.