Active inference for collective classification

Labeling nodes in a network is an important problem that has seen a growing interest. A number of methods that exploit both local and relational information have been developed for this task. Acquiring the labels for a few nodes at inference time can greatly improve the ac- curacy, however the question of figuring out which node labels to acquire is challenging. Previous approaches have been based on simple structural properties. Here, we present a novel technique, which we refer to as re- flect and correct, that can learn and predict when the un- derlying classification system is likely to make mistakes and it suggests acquisitions to correct those mistakes.