Efficient Global Learning of Entailment Graphs

Abstract

Entailment rules between predicates are fundamental to many semantic-inference applications. Consequently, learning such rules has been an active field of research in recent years. Methods for learning entailment rules between predicates that take into account dependencies between different rules (e.g., entailment is a transitive relation) have been shown to improve rule quality, but suffer from scalability issues, that is, the number of predicates handled is often quite small. In this article, we present methods for learning transitive graphs that contain tens of thousands of nodes, where nodes represent predicates and edges correspond to entailment rules (termed entailment graphs). Our methods are able to scale to a large number of predicates by exploiting structural properties of entailment graphs such as the fact that they exhibit a “tree-like” property. We apply our methods on two data sets and demonstrate that our methods find high-quality solutions faster than methods proposed in the past, and moreover our methods for the first time scale to large graphs containing 20,000 nodes and more than 100,000 edges.

Leveraging Linguistic Structure For Open Domain Information Extraction

“…Open IE triples have been used in a number of applications -for example, learning entailment graphs for new triples (Berant et al, 2011), and matrix factorization for unifying open IE and structured relations (Yao et al, 2012;Riedel et al, 2013). In each of these cases, the concise extractions provided by open IE allow for efficient symbolic methods for entailment, such as Markov logic networks or matrix factorization.…”

RELLY: Inferring Hypernym Relationships Between Relational Phrases

“…The comparison was performed on the task of creating entailment graphs as described in (Berant et al, 2011). This task is strongly related to finding hypernyms of relational phrases.…”

“…Moreover, the relations in the propositions were mainly limited to single verbs, whereas in our case we also consider longer relational phrases. Relations with semantic types were also used in typed entailment graphs (Berant et al, 2011). However, the type hierarchy was not considered there, which prevented from creating links between two relations with different semantic types.…”

Lexical Inference over Multi-Word Predicates: A Distributional Approach

“…Hagiwara et al (2009) identified synonyms using a supervised approach relying on distributional and syntactic features. Berant et al (2011) used distributional similarity between predicates to weight the edges of an entailment graph. By imposing global constraints on the structure of the graph, they obtained a more accurate set of inference rules.…”

“…Some filtered out frequent highly ambiguous verbs (Lewis and Steedman, 2013), others selected a single representative word (Melamud et al, 2013a), while yet others used multi-word LCs but treated them as fixed expressions (Lin and Pantel, 2001;Berant et al, 2011).…”