Decision support for personalized cloud service selection through multi-attribute trustworthiness evaluation

Area of Science:

• Computer Science
• Information Systems

Background:

• Growing demand for dependable and secure cloud services necessitates robust trustworthiness evaluation.
• Existing evaluation methods are complex, costly, and impractical for many customers, especially with limited historical data.
• Current models struggle with cloud services possessing few historical records.

Purpose of the Study:

• To propose a novel service selection approach that integrates missing value prediction and multi-attribute trustworthiness evaluation.
• To enable personalized trustworthy service selection using limited historical records.
• To improve upon existing methods for cloud service dependability and security assessment.

Main Methods:

• Developed a novel approach combining missing value prediction with multi-attribute trustworthiness evaluation.
• Utilized limited historical records for personalized service selection.
• Experimental validation against competing approaches.

Main Results:

• The proposed approach effectively handles cloud services with limited historical data.
• Experimental results demonstrate superior performance compared to existing methods.
• The approach aligns better with customer preferences and expectations for trustworthiness.

Conclusions:

• The novel approach provides a practical and personalized solution for cloud service selection.
• It enhances the ability to assess cloud service dependability and security effectively.
• This method overcomes limitations of existing trustworthiness evaluation techniques.