Constraint-Based Clustering Selection
This is an incremental improvement for researchers and practitioners in clustering, offering a simpler way to leverage supervision.
The paper tackles the problem of semi-supervised clustering by using pairwise constraints to select among clusterings from diverse unsupervised algorithms, rather than adapting individual algorithms, and shows that this approach often outperforms existing methods.
Semi-supervised clustering methods incorporate a limited amount of supervision into the clustering process. Typically, this supervision is provided by the user in the form of pairwise constraints. Existing methods use such constraints in one of the following ways: they adapt their clustering procedure, their similarity metric, or both. All of these approaches operate within the scope of individual clustering algorithms. In contrast, we propose to use constraints to choose between clusterings generated by very different unsupervised clustering algorithms, run with different parameter settings. We empirically show that this simple approach often outperforms existing semi-supervised clustering methods.