Cost-Effective Clustering through Active Feature-value Acquisition

Many datasets include feature values that are missing but may be acquired at a cost. In this paper, we consider the clustering task for such datasets, and address the problem of acquiring missing feature values that improve clustering quality in a cost-effective manner. Since acquiring all missing information may be unnecessarily expensive, we propose a framework for iteratively selecting feature values that result in highest improvements in clustering quality per unit cost. Our framework can be adapted to different clustering algorithms, and we illustrate it in the context of two popular methods, K-Means and hierarchical agglomerative clustering. Experimental results on several datasets demonstrate clustering accuracy improvements provided by the proposed framework over random acquisition. Additional experiments demonstrate the performance of the framework for different cost structures, and explore several alternative formulations of the acquisition strategy.