Semi-supervised clustering with inaccurate pairwise annotations

Pairwise relational information is a useful way of providing partial supervision in domains where class labels are difficult to acquire. This work presents a clustering model that incorporates pairwise annotations in the form of and relations and considers possible annotation inaccuracies (i.e., a common setting when and relations generated by stochastic block models. We adopt a maximum-likelihood approach and demonstrate that, even when supervision is weak and inaccurate, accounting for relational information significantly improves clustering performance. We observe that relational information also helps to detect meaningful groups in real-world datasets that do not fit the original data-distribution assumptions. Additionally, we extend the model to integrate prior knowledge of experts’ accuracy and discuss circumstances in which the use of this knowledge is beneficial.