A multi-class approach for ranking graph nodes: Models and experiments with incomplete data

After the phenomenal success of the PageRank algorithm, many researchers have extended the PageRank approach to ranking graphs with richer structures in addition to the simple linkage structure. Indeed, in some scenarios we have to deal with networks modeling multi-parameters data where each node has additional features and there are important relationships between such features. This paper addresses the need of a systematic approach to deal with multi-parameter data. We propose models and ranking algorithms that can be applied to a large variety of networks (bibliographic data, patent data, twitter and social data, healthcare data). We focus on several aspects not previously addressed in the literature: (1) we propose different models for ranking multi-parameters data and a class of numerical algorithms for efficiently computing the ranking score of such models, (2) we analyze stability and convergence of the proposed numerical schemes and we derive a fast and stable ranking algorithm, (3) we analyze the robustness of our models when data are incomplete. The comparison of the rank on the incomplete data with the rank on the full structure shows that our models compute consistent rankings whose correlation is up to 60% when just 10% of the links of the attributes are maintained.