2008
Recent developments in deep linguistic processing give rise to renewed optimism concerning the practical applicability of advanced grammatical analysis. For HPSG parsing a breakthrough in efficiency was achieved through a new form of international collaboration that lead to improved combinations of methods in a rather systematic way. Yet efficiency is just one of the obstacles to the utilization of deep processing in real life applications. A novel approach to the exploitation of deep parsing offers a strategy for compensating the deficiency in coverage and robustness. Through a combination of deep and shallow processing, the robustness of shallow processing is preserved for information extraction applications that exploit parsing methods of different depth. The underlying strategy and the realized architecture also provide a basis for distributed collective research on novel combinations of processing methods including applications of hyper learning.
Cited by 0
Hans Uszkoreit · 2002
Recent developments in deep linguistic processing give rise to renewed optimism concerning the practical applicability of advanced grammatical analysis. For HPSG parsing a breakthrough in efficiency was achieved through a new form of international collaboration that lead to improved combinations of methods in a rather systematic way. Yet efficiency is just one of the obstacles to the utilization of deep processing in real life applications. A novel approach to the exploitation of deep parsing offers a strategy for compensating the deficiency in coverage and robustness. Through a combination of deep and shallow processing, the robustness of shallow processing is preserved for information extraction applications that exploit parsing methods of different depth. The underlying strategy and the realized architecture also provide a basis for distributed collective research on novel combinations of processing methods including applications of hyper learning.
Cited by 26
Francisco Nuno Quintiliano Mendonça Carapeto Costa · Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT) · 2007
Tese de mestrado em Informática, apresentada à Universidade de Lisboa, através da Faculdade de Ciências, 2007
Cited by 2
Dan Flickinger, Yi Zhang, Valia Kordoni · 2010
Cited by 0
Aleš Horák, Karel Pala, Marie Duží et al. · 2007
In the paper, we describe methods for exploitation of a new lexical database of valency frames (VerbaLex) in relation to Transparent Intensional Logic (TIL). We present a detailed description of the Complex Valency Frames (CVF) as they appear in VerbaLex including basic ontology of the VerbaLex semantic roles.
Cited by 2Open access
Paris Rocquencourt · 2010
Cited by 0
Baldwin Timothy · Waseda University Repository (Waseda University) · 2007
Coverage has been a constant thorn in the side of deployed deep linguistic processing applications, largely because of the difficulty in constructing, maintaining and domaintuning the complex lexicons that they rely on.This paper reviews various strands of research on deep lexical acquisition (DLA), i.e. the (semi-)automatic creation of linguistically-rich language resources, particularly from the viewpoint of DLA for precision grammars.
Cited by 0Open access
Gerold Schneider · Zurich Open Repository and Archive (University of Zurich) · 2004
This paper describes Pro3Gres, a fast, robust, broad-coverage parser that delivers deep-linguistic grammatical relation structures as output, which are closer to predicate-argument structures and more informative than pure constituency structures. The parser stays as shallow as is possible for each task, combining shallow and deep-linguistic methods by integrating chunking and by expressing the majority of long-distance dependencies in a context-free way. It combines statistical and rule-based approaches, different linguistic grammar theories and different linguistic resources. Preliminary evaluations indicate that the parsers performance is state-of-the-art.
Cited by 3Open access
2007
Cited by 2
António Branco, Francisco Costa · 2008
This paper proposes a design strategy for deep language processing grammars to appropriately handle language variants. It allows a grammar to be restricted as to what language variant it is tuned to, but also to detect the variant a given input pertains to. This is evaluated and compared to results obtained with an alternative strategy by which the relevant variant is detected with current language identification methods in a preprocessing step. 1
Cited by 0
Francisco Nuno, Quintiliano Mendonça, Carapeto Costa et al. · 2007
Cited by 0
Francisco Costa, António Branco · Journal of Language Modelling · 2013
The full-fledged processing of temporal information presents specific challenges. These difficulties largely stem from the fact that the temporal meaning conveyed by grammatical means interacts with many extra-linguistic factors (world knowledge, causality, calendar systems, reasoning). This article proposes a novel approach to this problem, based on a hybrid strategy that explores the complementarity of the symbolic and probabilistic methods. A specialized temporal extraction system is combined with a deep linguistic processing grammar. The temporal extraction system extracts eventualities, times and dates mentioned in text, and also temporal relations between them, in line with the tasks of the recent TempEval challenges; and uses machine learning techniques to draw from different sources of information (grammatical and extra-grammatical) even if it is not explicitly known how these combine to produce the final temporal meaning being expressed. In turn, the deep computational grammar delivers richer truth-conditional meaning representations of input sentences, which include a principled representation of temporal information, on which higher level tasks, including reasoning, can be based. These deep semantic representations are extended and improved according to the output of the aforementioned temporal extraction module. The prototype implemented shows performance results that increase the quality of the temporal meaning representations and are better than the performance of each of the two components in isolation.
Cited by 1Open access
Lynott, Dermot · Research Explorer (The University of Manchester) · 2012
Cited by 0
Louise Connell, Dermot Lynott · Psychonomic Bulletin & Review · 2013
Cited by 82Open access
Branco, António Horta, Costa, Francisco · Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT) · 2008
This document presents the implementation of LXGram, in its version A.4.1. LXGram is a grammar for the computational processing of Portuguese being developed at the University of Lisbon. The present document has TR-2008-16 as its companion document, which provides a methodological introduction to the computational grammar presented here. Though they can be used independently of each other, these two documents should be seen as two parts of the same text. The present document was written in English so that the results reported in it may be used by the widest research community. The document TR-2008-16, in turn, was written in Portuguese in order for its content to be made available, for the firrst time, for an audience of speakers of Portuguese --- as other texts in English exist with approximate goals or coverage
Cited by 3Open access
Kostadin Cholakov, Valia Kordoni, Yi Zhang · 2008
In this paper we illustrate and underline the importance of making detailed linguistic information a central part of the process of automatic acquisition of large-scale lexicons as a means for enhancing robustness and at the same time ensuring maintainability and re-usability of deep lexicalised grammars. Using the error mining techniques proposed in (van Noord, 2004) we show very convincingly that the main hindrance to portability of deep lexicalised grammars to domains other than the ones originally developed in, as well as to robustness of systems using such grammars is low lexical coverage. To this effect, we develop linguistically-driven methods that use detailed morphosyntactic information to automatically enhance the performance of deep lexicalised grammars maintaining at the same time their usually already achieved high linguistic quality. 1
Cited by 0
Dop, Thomas
Cited by 0
JoAnn P. Silkes, Carolyn Baker, Tracy Love · Topics in Language Disorders · 2020
This study investigates learning in aphasia as manifested through automatic priming effects. There is growing evidence that people with aphasia have impairments beyond language processing that could affect their response to treatment. Therefore, better understanding these mechanisms would be beneficial for improving methods of rehabilitation. This study assesses semantic and repetition priming effects at varied interstimulus intervals, using stimuli that are both non-linguistic and linguistic in tasks that range from requiring nearly no linguistic processing to requiring both lexical and semantic processing. Results indicate that people with aphasia maintain typical patterns of learning across both linguistic and non-linguistic tasks as long as the implicit prime-target relationship does not depend on deep levels of linguistic processing. As linguistic processing demands increase, those with agrammatic aphasia may require more time to take advantage of learning through implicit prime-target relationships, and people with both agrammatic and non-agrammatic aphasia are more susceptible to breakdown of the semantic networks as processing demands on that system increase.
Cited by 16Open access
Xue, Yuyan, Williams, John · eScholarship (California Digital Library) · 2023
Current theories suggest that linguistic relativity and language embodiment are two sides of the same coin and that language embodiment and “linguistic shortcut” are two alternative types of semantic processing. It can therefore be deduced that deep embodied simulation of linguistic labels may facilitate the corresponding linguistic relativity effect more than shallow linguistic-shortcut processing does. However, this theoretical deduction has not been tested empirically. Here, participants used two newly-trained labels of verb (in)transitivity in sentence contexts that were either simulated in a deep, embodied manner (visual imagery group) or processed via shallow linguistic-shortcut (linguistic associate group). Subsequently, the former group were more likely to categorize motion events based on motion transitivity than the latter group. We thus expanded the moderators of linguistic relativity to the types of semantic access. The findings are discussed in light of the relativist, the universalist, and the sociocultural view on language-cognition relationship.
Cited by 0Open access
Salvatore Romeo, Giovanni Da San Martino, Alberto Barrón‐Cedeño et al. · 2018
Although deep neural networks have been proving to be excellent tools to deliver state-of-the-art results, when data is scarce and the tackled tasks involve complex semantic inference, deep linguistic processing and traditional structure-based approaches, such as tree kernel methods, are an alternative solution. Community Question Answering is a research area that benefits from deep linguistic analysis to improve the experience of the community of forum users. In this paper, we present a UIMA framework to distribute the computation of cQA tasks over computer clusters such that traditional systems can scale to large datasets and deliver fast processing.
Cited by 1Open access