FACTORIE: Efficient Probabilistic Programming for Relational Factor Graphs via Imperative Declarations of Structure, Inference and Learning
Andrew McCallum, Khashayar Rohanimanesh, Michael Wick, Karl Schultz, Sameer Kumar Singh · ScholarWorks@UMassAmherst (University of Massachusetts Amherst) · 2008
Discriminatively trained undirected graphical models have garnered tremendous interest and empirical success in natural language processing, computer vision, bioinformatics and many other areas [16,1,11].Some of these models use simple structure (e.g.linear-chains, grids, fully-connected affinity graphs), but there has been increasing interest in more complex relational structurecapturing more arbitrary dependencies among sets of variables, in repeated patterns.Reimplementing variant structures from scratch is difficult and error-prone, however, and thus there have been several efforts to provide a high-level language in which new undirected model structures can be specified.These include SQL [17], first-order logic [13], and others such as Csoft [18].