Inducing Semantic Representation from Text by Jointly Predicting and Factorizing Relations
This work addresses the challenge of unsupervised semantic role induction for natural language processing, representing an incremental improvement by integrating existing techniques.
The authors tackled the problem of inducing semantic representations from text by jointly predicting and factorizing relations, achieving performance on par with top role induction methods on English without using prior linguistic knowledge.
In this work, we propose a new method to integrate two recent lines of work: unsupervised induction of shallow semantics (e.g., semantic roles) and factorization of relations in text and knowledge bases. Our model consists of two components: (1) an encoding component: a semantic role labeling model which predicts roles given a rich set of syntactic and lexical features; (2) a reconstruction component: a tensor factorization model which relies on roles to predict argument fillers. When the components are estimated jointly to minimize errors in argument reconstruction, the induced roles largely correspond to roles defined in annotated resources. Our method performs on par with most accurate role induction methods on English, even though, unlike these previous approaches, we do not incorporate any prior linguistic knowledge about the language.