Automatic fine-grained semantic classification for domain adaptation

Abstract

Assigning arguments of verbs to different semantic classes ('semantic typing'), or alternatively, checking the 'selectional restrictions' of predicates, is a fundamental component of many natural language processing tasks. However, a common experience has been that general purpose semantic classes, such as those encoded in resources like WordNet, or handcrafted subject-specific ontologies, are seldom quite right when it comes to analysing texts from a particular domain. In this paper we describe a method of automatically deriving fine-grained, domain-specific semantic classes of arguments while simultaneously clustering verbs into semantically meaningful groups: the first step in verb sense induction. We show that in a small pilot study on new examples from the same domain we are able to achieve almost perfect recall and reasonably high precision in the semantic typing of verb arguments in these texts.

A social dimensional cyber threat model with formal concept analysis and fact-proposition inference

“…The approach combines rough set theory and concept lattice theory to measure the concept nodes from two ontologies based on Tversky's similarity model. The use of an overlap coefficient to obtain similarity measure has also been used in the works of Liakata and Pulman (2008) and Bhagat et al (2007). Hospodka (2008) used Fact Proposition Space Inference to provide a valuable information fusion and belief integration engine.…”

Deriving generalized knowledge from corpora using WordNet abstraction

“…Minimizing word sense ambiguity by focusing on a specific domain was later seen in the work of Liakata and Pulman (2008), who performed hierarchical clustering using output from their KNEXT-like system first described in (Liakata and Pulman, 2002). Terminal nodes of the resultant structure were used as the basis for inferring semantic type restrictions, reminiscent of the use of CBC clusters (Pantel and Lin, 2002) by Pantel et al (2007), for typing the arguments of paraphrase rules.…”

Building a semantic lexicon of English nouns via bootstrapping

“…Roark and Charniak (1998) constructed a semantic lexicon using co-occurrence statistics of nouns within noun phrases. More recently, Liakata and Pulman (2008) induced a hierarchy over nominals using as features knowledge fragments similar to the sort given by KNEXT. Our work might be viewed as aiming for the same goal (a lexico-semantic based partitioning over nominals, tied to corpus-based knowledge), but allowing for an a priori bias regarding preferred structure.…”