Evaluating Social Networks Using Task-Focused Network Inference
This work addresses the challenge of selecting appropriate network representations for social behavior modeling, but it appears incremental as it builds on existing network inference and evaluation methods.
The authors tackled the problem of evaluating how well given social networks model specific behaviors by introducing a framework to compare them against data-inferred networks for predictive tasks, applying it to a Last.fm dataset for music preference classification.
Networks are representations of complex underlying social processes. However, the same given network may be more suitable to model one behavior of individuals than another. In many cases, aggregate population models may be more effective than modeling on the network. We present a general framework for evaluating the suitability of given networks for a set of predictive tasks of interest, compared against alternative, networks inferred from data. We present several interpretable network models and measures for our comparison. We apply this general framework to the case study on collective classification of music preferences in a newly available dataset of the Last.fm social network.