Faceted Ranking of Egos in Collaborative Tagging Systems

TL;DR

This paper introduces a scalable method for computing personalized user rankings in folksonomies by combining offline tag-based PageRank scores with an efficient online merging process for multiple tags.

Contribution

It proposes a novel faceted ranking approach using PageRank, enabling real-time ego-ranking in collaborative tagging systems.

Findings

Offline tag-based PageRank is scalable.

Online merging algorithm has linear time complexity.

Faceted ranking improves user reputation evaluation.

Abstract

Multimedia uploaded content is tagged and recommended by users of collaborative systems, resulting in informal classifications also known as folksonomies. Faceted web ranking has been proved a reasonable alternative to a single ranking which does not take into account a personalized context. In this paper we analyze the online computation of rankings of users associated to facets made up of multiple tags. Possible applications are user reputation evaluation (ego-ranking) and improvement of content quality in case of retrieval. We propose a solution based on PageRank as centrality measure: (i) a ranking for each tag is computed offline on the basis of the corresponding tag-dependent subgraph; (ii) a faceted order is generated by merging rankings corresponding to all the tags in the facet. The fundamental assumption, validated by empirical observations, is that step (i) is scalable. We also present algorithms for part (ii) having time complexity O(k), where k is the number of tags in the facet, well suited to online computation.