Semi-supervised blockmodelling with pairwise guidance

Abstract

© 2019, Springer Nature Switzerland AG. Blockmodelling is an important technique for detecting underlying patterns in graphs. Existing blockmodelling algorithms are unsupervised and cannot take advantage of the existing information that might be available about objects that are known to be similar. This background information can help finding complex patterns, such as hierarchical or ring blockmodel structures, which are difficult for traditional blockmodelling algorithms to detect. In this paper, we propose a new semi-supervised framework for blockmodelling, which allows background information to be incorporated in the form of pairwise membership information. Our proposed framework is based..