2015 | OriginalPaper | Buchkapitel
SimRank Based Top-k Query Aggregation for Multi-Relational Networks
verfasst von : Jing Xu, Cuiping Li, Hong Chen, Hui Sun
Erschienen in: Web-Age Information Management
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
SimRank is one measure that compute the similarities between nodes in applications, where the returning of top-k query lists is often required. In this paper, we adopt SimRank as the similarity computation measure and re-write the original inefficient iterative equation into a non-iterative one, we call it Eigen-SimRank. We focus on multi-relational networks, where there may exist different kinds of relationships among nodes and query results may change with different perspectives. In order to compute a top-k query list under any perspective especially compound perspective, we suggest dynamic updating algorithm and rank aggregation methods. We evaluate our algorithms in the experiment section.