Structuring relations for fusion in intelligence

Humans daily infer causal structure from patterns of correlation and learn about categories and hidden properties of objects based on experience and knowledge. A Bayesian approach seems to best model human reasoning over structures, relations and links, and it is possible to provide a detailed computational account of how a number of basic structural forms can be inferred from various types of data (feature sets, similarity matrices, relations). In the literature some algorithms have been proposed that generate candidate model structures from graph grammars, compute the probability of the data given each candidate model, and identify the model with maximum posterior probability given the data. The structural representation, being generated by algorithms comparable to human thinking (according to the cognitive sciences community) should be also easily understandable and usable by analysts for further investigation.