Question answering from structured knowledge sources

Abstract

We present an implemented approach for domain-restricted question answering from structured knowledge sources, based on robust semantic analysis in a hybrid NLP system architecture. We perform question interpretation and answer extraction in an architecture that builds on a lexical-conceptual structure for question interpretation, which is interfaced with domain-specific concepts and properties in a structured knowledge base. Question interpretation involves a limited amount of domain-specific inferences, and accounts for higher-level quantificational questions. Question interpretation and answer extraction are modular components that interact in clearly defined ways. We derive so-called proto queries from the linguistic representations, which provide partial constraints for answer extraction from the underlying knowledge sources. The search queries we construct from proto queries effectively compute minimal spanning trees from the underlying knowledge sources. Our approach naturally extends to multilingual question answering, and has been developed as a prototype system for two application domains: the domain of Nobel prize winners, and the domain of Language Technology, on the basis of the large ontology underlying the information portal LT WORLD.

79. Electronic dictionaries for question answering and natural language database access

“…The QA system of Bouma et al (2005), for instance, uses an HPSGbased parser that returns dependency tree representations. Most existing grammar-based QA approaches aim at building shallow or underspecified logical representations, typically by means of elementary, lemma-based predicates associated with the entries of the underlying lexicon (Bos 2006;Frank et al 2007;Bobrow et al 2007; see Section 2.2). The lexical semantic representations of these approaches often do not account for word sense distinctions, which is problematic for drawing inferences, be it simple hypernym generalization or more advanced logical reasoning based on axiomatized knowledge.…”

“…At present, these specifications are essentially word-based predicates of appropriate arity without sense distinctions (except for the possibly disjunctive assignment of ontological sorts in the German lexicon). Both lexicons are used in the multilingual NLADB architecture of Frank et al (2007) for constructing an MRS representation of the user question. A conceptual representation is then generated in a post-processing step that involves semantic LRs including WordNet and FrameNet (see Section 3).…”

“…Furthermore, Cyc has been used to enrich the XLE lexicon with conceptual information ( Moldovan et al (2007), which is employed for answer validation. Another application is described in Frank et al (2007), where SUMO is used as the common reference ontology of an NLADB system (see below).…”

How Useful Are Natural Language Interfaces to the Semantic Web for Casual End-Users?

“…Considering the difficulties with full NL, it seems comprehensible that restricted NL or menuguided interfaces have been proposed by some approaches [14,22,25]. The popularity of the Semantic Web created a number of NLIs that provide access to ontology-based knowledge bases [9,11,12,15,18,26]. Most of the evaluations of NLIs mainly focus on retrieval performance and/or the portability dimension.…”

“…Several projects have shown that NLIs can perform well in retrieval tasks [12,20,24] and be portable as well as domain-independent [9,18,26] without being unnecessarily complex. This paper now attempts to shed some light on the problem dimension of usability and usefulness of NLIs (i.e., the last of the four issues raised above).…”