A Trust Model for Ubiquitous Healthcare Environment on the Basis of Adaptable Fuzzy-Probabilistic Inference System.

Area of Science:

• Computer Science
• Healthcare Informatics
• Artificial Intelligence

Background:

• Trust is crucial for patient satisfaction in psychologist selection within ubiquitous healthcare (UH).
• Anxiety disorder patients require reliable healthcare provider selection.
• Existing trust models lack robustness in dynamic healthcare environments.

Purpose of the Study:

• To propose and validate an adaptable cloud inference system (ACIS) for estimating psychologist trustworthiness in UH.
• To enhance patient satisfaction by ensuring reliable psychologist selection.
• To improve the accuracy and robustness of trust value (TV) estimation.

Main Methods:

• Development of an ACIS using cloud theory and fuzzy-probabilistic reasoning.
• Incorporation of a soft AND cloud relation operator tuned by information source consistency.
• Validation through theoretical analysis, MATLAB simulations, and experimental investigations.

Main Results:

• ACIS achieved 45.5% higher accuracy than state-of-the-art models in clean environments.
• ACIS demonstrated 62% higher accuracy under large-scale collusive malicious attacks.
• High true positive rate (0.9) in discriminating untrustworthy providers, even with 40% erroneous reputation data.

Conclusions:

• The proposed ACIS effectively estimates psychologist trustworthiness in UH settings.
• The ACIS enhances robustness against erroneous data and malicious attacks.
• Implementing this trust model significantly contributes to achieving patient satisfaction in UH.