Jason: A Scalable Reputation System for the Semantic Web

Abstract

Abstract. The recent development of the Internet, especially the expanding use of social software and dynamic content generation commonly termed as Web 2.0 enables users to find information about almost every possible topic on the Web. On the downside, it becomes more and more difficult to decide which information can be trusted in. In this paper we propose the enhancement of Web 2.0 by a scalable and secure cross-platform reputation system that takes into account a user's social network. Our proposed solution Jason is based on standard methods of the semantic web and does not need a central entity. It enables the fast and flexible evaluation of arbitrary content on the World Wide Web. In contrast to many other reputation systems it provides mechanisms to ensure the authenticity of web content, thus, enabling the user to explicitely choose information published by trusted authors.

Pengembangan Model Pengukuran Sistem Informasi Akademik Terhadap Kepuasan Pengguna Menggunakan Technology Acceptance Model Dan Delone&Mclean

“…Pada tahapan penelitian gambar 1 diawali dengan tahap 1 yaitu survey sistem akademik berbasis web yang akan dijadikan sebagai objek dari analisis penelitian, kemudian membuat pengembangan model yang akan diusulkan, dilanjutkan dengan pengambilan sampel untuk penilaian dari model yang diusulkan. [10] Selanjutnya pengumpulan data dilakukan dan dilakukan analisis pada model yang diusulkan untuk mengetahui apakah data ini valid dan reabilitas terhadap model.…”

The Need for Interoperable Reputation Systems

“…Here is the need for solutions that can be easily integrated in the existing web technologies. The suggestion we implemented in [19] was to use the Resource Description Framework (RDF) [13] common for the Web 2.0 and allowing to add reputation information as meta-information to arbitrary web content. -Algorithm: In every reputation system different implementations of rating function, evaluation function and reputation function are defined depending on the system designer.…”

Privacy and Liveliness for Reputation Systems

“…A central problem for privacy-respecting reputation systems is that they must guarantee that users cannot abolish negative reputation. This can either be reached by only allowing positive reputation, as proposed by Voss and Androulaki et al [5,6], by making it difficult for the user to distinguish between positive and negative ratings, as proposed in [7] by Steinbrecher, or by a trusted third party. Thereby, this trusted third party can either be an external reputation provider, as proposed by Pingel et al and Anwar et al [8,9] or a trusted platform module for the user, as proposed in [10,5] by Kinateder et al and Voss. Anonymity of the users involved is not as easy as just using anonymizing services on the network layer.…”

“…Furthermore, the number of reputation values is fixed and small. Even if we do not specify a concrete reputation function here, this requirement must be met in order to restrict identifiability of users by their reputation values, see requirement (7). All users will finally have the best reputation value and will keep it, thus the system reaches a final state and becomes useless, i.e., no user has an incentive to behave fairly, which violates fairness (5) and liveliness of reputation (4).…”