Masterarbeit
Migrating Fuzzy Logic-based Trust Models to the Redecentralized Web
Completion
2024/07
Research Area
Students
Emon Hossain
Shuvo Kumar Sarkar
Advisers
Description
The evolution of the World Wide Web has been remarkable, becoming essential in our everyday lives. In the existing centralized web framework, we, the users, lack substantial control over our data. It's unclear who accesses our data and where it's stored, forcing us to depend heavily on central authorities like major tech companies. Initiatives like the European Union's Next Generation Internet and Tim Berners-Lee's Social Linked Data (SoLiD) are advocating for this shift back to web redecentralization. This move towards redecentralization aims to enhance data privacy and reduce dependence on centralized entities. In the existing centralized web framework, both consumers and providers depend on central authorities to form trust relationships and make trust-based decisions for them.
The redecentralized web can be modelled as an open and dynamic Multi-Agent Systems (MAS). The dynamic aspect of MAS is highlighted by the agents' ability to join or leave the system at any time, operating independently in their decision-making. In the context of a redecentralized web, a web application may rely on other applications to attain its goals. Key questions emerge in these interactions: Who is trustable? Why and how can a web application trust other web applications? Addressing these questions highlights the importance of trust and reputation in the redecentralized web. Therefore, managing trust and reputation within a redecentralized web is crucial to minimize malicious or harmful data usages. In the context of MAS, several trust management models have been proposed to facilitate the possibility of one agent being taken advantage of by others. These models assist agents in evaluating the trustworthiness of other agents, and the agent is then able to make decisions that are worthy of their trust. In the agent's trustworthiness research domain, the "Hybrid Trust Model (HTM)" and the "Fuzzy logic and Q-learning based trust model (FQT)" are potential trust models to evaluate trust autonomously and effectively. The objective of this team thesis is to migrate the HTM model and the FQT model to a redecentralized web so that it can help web applications make trust-aware decisions autonomously and examine how this adaptation works for the redecentralized web with the given use case of linked data acquisition of one web application.
The objective of this master thesis is to find an approach or a combination of approaches to solve the mentioned problem for the migration of fuzzy logic-based trust models to the redecentralized web. This particularly includes the state of the art regarding trust models and their migration to new use cases. The demonstration of feasibility with an implementation prototype of the concept is part of this thesis as well as a suitable evaluation with exemplary and empirical use cases.
