Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
18625 papers
The goal of the current paper is to describe the TLGbank, a treebank of type-logical proof semi-automatically extracted from the French Treebank. Though the framework chosen for the treebank are multimodal type-logical grammars, we have ensured that the analysis is compatible with other mondern type-logical grammars, such the displacement calculus and first-order linear logic. We describe the extraction procedure, analyse first results and compare the treebank to the CCGbank.
Research into statistical parsing for English has enjoyed over a decade of successful results. However, adapting these models to other languages has met with difficulties. Previous comparative work has shown that Modern Arabic is one of the most difficult languages to parse due to rich morphology and free word order. Classical Arabic is the ancient form of Arabic, and is understudied in computational linguistics, relative to its worldwide reach as the language of the Quran. The thesis is based on seven publications that make significant contributions to knowledge relating to annotating and parsing Classical Arabic. A central argument of this thesis is that using a hybrid representation closely aligned to traditional grammar leads to improved parsing for Arabic. To test this hypothesis, two approaches are compared. As a reference, a pure dependency parser is adapted using graph transformations, resulting in an 87.47% F1-score. This is compared to an integrated parsing model with an F1-score of 89.03%, demonstrating that joint dependency-constituency parsing is better suited to Classical Arabic.
A Treebank is a linguistic resource that is composed of a large collection of manually annotated and verified syntactically analyzed sentences. Statistical Natural Language Processing (NLP) approaches have been successful in using these annotations for developing basic NLP tasks such as tokenization, diacritization, part-of-speech tagging, parsing, among others. In this paper, we address the problem of exploiting Treebank resources for statistical parsing of Modern Standard Arabic (MSA) sentences. Statistical parsing is significant for NLP tasks that use parsed text as an input such as Information Retrieval, and Machine Translation. We conducted an experiment on Pen Arabic Treebank (PATB) and the parsing performance obtained in terms of Precision, Recall, and F-measure was 82.4%, 86.6%, 84.4%, respectively.
This demonstration presents a freely available open source lexical database omorfi. Omorfi is a mature lexicographical database project, started out as a single-person single-purpose free open source morphological anal-yser project, omorfi has since grown to be used in variety of applications including spell-checking, statistical and rule-based machine translation, treebanking, joint syntactic and morphological parsing, poetry generation, information extraction. In this demonstration we hope to show both the variety of end-user facing applications as well as the tools and interfaces for computational linguists to make the best use of a developing product. We show a shallow database arrangement that has allowed a great variety of contributors from different projects to extend the lexical database while not breaking the continued use of existing end-applications. We hope to show both the best current practices for lexical data management and software engineering with regards to continuous external project integra-tion of a constantly developing product. As case examples we show some of the integrations with following applications: Voikko spell-checking for Windows, Mac OS X, Linux and Android, statistical machine translation pipelines with moses, rule-based machine translation with apertium and traditional xerox style morphological analysis and generation. morpho-logical segmentation, as well as application programming interfaces for python and Java. 1
Liang-Chih Yu, Jin Wang, K. Robert Lai, Xue-jie Zhang. Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Short Papers). 2015.
Abstract. This paper proposes a hierarchical model to parse both En-glish and Chinese sentences. This is done by iteratively constructing simple constituents first, so that complex ones could be detected reliably with richer contextual information in the following processes. Evalua-tion on the Penn WSJ Treebank and the Penn Chinese Treebank using maximum entropy models shows that our method can achieve a good performance with more flexibility for future improvement.
For many students, coming to learn mathematics is as much about the pedagogical relay through which concepts are conveyed as it is about the mathematics per se. This relay comprises social, cultural and linguistic norms as well as the mathematical discourse. In this study, I outline the practices of one remote school and how the teaching practices scaffold Indigenous learners whose home language is different from the language of instruction (Standard Australian English) as they come to learn mathematics. Through various strategies, teachers have created positive learning environments that celebrate the home languages of the students while supporting the transition into Standard Australian English and the discourse of mathematics.
Individually, both near-misses and losses disguised as wins (LDWs) have been seen to exert pro-motivational effects on gambling. However, it is not clear whether both structural characteristics are effective within the same game. Participants (n = 40) played a slot machine simulation. The simulation delivered near-misses, wins and ‘full-misses’. Half the participants also received LDWs that occurred independently of the outcomes on the payline. Valence and motivation ratings were collected after each round. Results showed that the LDW group reported increased valence ratings compared to the no-LDW group. Within the LDW group, trials with LDWs also resulted in increased enjoyment compared to trials without LDWs. We distinguished near-misses falling either side of the payline. Near-misses before the payline (NMB) were rated as more motivational than near-misses after the payline (NMA), whereas NMAs were rated as more aversive than NMBs. These differences between the two near-miss types were exacerbated by LDWs. Results demonstrate LDWs increase the trial-by-trial enjoyment of non-win outcomes. The motivational and hedonic effects of near-misses differed for events either side of the payline, and these differences were exaggerated by the presence of LDWs. Thus, near-misses can retain their effectiveness in complex forms of gambling that also deliver LDWs.
Earlier studies have demonstrated emotional overreactions to affective visual stimuli in patients with borderline personality disorder (BPD). However, contradictory findings regarding hyper- versus hyporeactivity have been reported for peripheral physiological measures. In order to extend previous results, the authors investigated emotional reactivity and long-term habituation in the acoustic modality. Twenty-two female BPD patients and 19 female nonclinical controls listened to emotionally negative, neutral, and positive sounds in two identical sessions. Heart rate, skin conductance, zygomaticus/corrugator muscle, and self-reported valence/arousal responses were measured. BPD patients showed weaker skin conductance responses to negative sounds than controls. The elevated zygomaticus activity in response to positive sounds observed in controls was absent in BPD patients, and BPD patients assigned lower valence ratings to positive sounds than controls. In Session 2, patients recognized fewer positive sounds than controls. Across both groups, physiological measures habituated between sessions. These findings add to growing evidence toward partial affective hyporeactivity in BPD.
We study non-deterministic oracles for training non-projective beam search parsers with swap transitions. We map out the spurious ambiguities of the transition system and present two non-deterministic oracles as well as a static oracle that minimizes the number of swaps. An evaluation on 10 treebanks reveals that the difference between static and non-deterministic oracles is generally insignificant for beam search parsers but that non-deterministic oracles can improve the accuracy of greedy parsers that use swap transitions.
International audience
International audience
International audience
Presentations given at the workshop Language Comparison with Linguistic Databases (LanCLiD 2), held on 30 April 2015 at the Department of Linguistics of the Max Planck Institute for Evolutionary Anthropology (Leipzig, Germany).
Graph-based dependency parsing algorithms commonly employ features up to third order in an attempt to capture richer syntactic relations. However, each level and each feature combination must be defined manually. Besides that, input features are usually represented as huge, sparse binary vectors, offering limited generalization. In this work, \nwe present a deep architecture for dependency parsing based on a convolutional neural network. It can examine the whole sentence structure before scoring each head/modifier \ncandidate pair, and uses dense embeddings as input. Our model is still under ongoing work, achieving 91.6% unlabeled attachment score in the Penn Treebank.
The paper reports experiences of automatically converting the dependency analysis of the LinES English-Swedish parallel treebank to universal dependencies (UD). The most tangible result is a version of the treebank that actually employs the relations and parts-of-speech categories required by UD, and no other. It is also more complete in that punctuation marks have received dependencies, which is not the case in the original version. We discuss our method in the light of problems that arise from the desire to keep the syntactic analyses of a parallel treebank internally consistent, while available monolingual UD treebanks for English and Swedish diverge somewhat in their use of UD annotations. Finally, we compare the output from the conversion program with the existing UD treebanks.
In this paper, I discuss the theoretical and practical issues raised in the development of Tibidabo, a syntactically and semantically annotated corpus of Spanish, which has been developed using the DELPH-IN corpus annotation environment and the HPSG-based grammar, SRG. I describe in detail the linguistic analysis that Tibidabo provides, focussing on various syntactic phenomena that are frequently discussed in the literature and consider alternative analyses proposed in the HPSG literature and in AnCora, another Spanish Treebank. 2 2 This work was funded by the Ramón y Cajal programme of the Spanish Ministerio de Ciencia e Innovación.
This paper presents a resource and the associated annotation process used in a project of interlinking Czech and English verbal translational equivalents based on a parallel, richly annotated dependency treebank containing also valency and semantic roles, namely the Prague Czech-English Dependency Treebank. One of the main aims of this project is to create a high-quality and relatively large empirical base which could be used both for linguistic comparative research as well as for natural language processing applications, such as machine translation or cross-language sense disambiguation. This paper describes the resulting lexicon, CzEngVallex, and the process of building it, as well some interesting observations and statistics already obtained.
This is the annotated corpus developed for Afrikaans for the Afribooms project. The corpus includes annotations for lemma, part-of-speech (POS) and dependency relations. Lemma and POS information originates from the source corpus used, in this project only the dependency tags and relations were added.
We present a syntactic analysis query toolkit geared specifically towards massive dependency parsebanks and morphologically rich languages. The query language allows arbitrary tree queries, including negated branches, and is suitable for querying analyses with rich morphological annotation. Treebanks of over a million words can be comfortably queried on a low-end netbook, and a parsebank with over 100M words on a single consumer-grade server. We also introduce a web-based interface for interactive querying. All contributions are available under open licenses.
Syntactic-semantic treebank for domain ontology creationThis paper focuses on the creation of a domain treebank for the purposes of compiling a domain ontology. The domain treebank is viewed as a suitable resource for extracting of semantic relations from syntactic structures. First, the steps for ontology building are considered. Then, the processing over glossaries and standards is described with regard to their syntactic annotation. The utility of deriving semantic knowledge from the Treebank is also illustrated via the basic phrases. The idea is that the domain knowledge is represented in the domain data, but via treebanking more linguistic patterns can be extracted, which to be mapped to concepts and relations in a domain ontology.
In this report we present the results obtained analysing the use, frequency of use and the position of adverbial clauses. This analysis has been performed in the Basque Dependency Treebank (BDT). We also have used the descriptive grammars of Euskaltzaindia, the Royal Academy of the Basque.
We highlight the main changes recently undergone by the Italian Dependency Treebank in the transition to an extended and revised edition, compliant with the annotation schema of Universal Dependencies. We explore how these changes affect the accuracy of dependency parsers, performing comparative tests on various versions of the treebank. Despite significant changes in the annotation style, statistical parsers seem to cope well and mostly improve.
Alexithymia is believed to involve deficits in emotion processing and imagery ability. Previous findings suggest that it is especially related to deficits in processing the arousal dimension of emotion, and that discordance may exist between self-report and physiological responses to emotional stimuli in alexithymia. The current study used a well-established emotional imagery paradigm to examine emotion processing deficits and discordance in participants (N = 86) selected based on their extreme scores on the Toronto Alexithymia Scale-20. Physiological (skin conductance, heart rate, and corrugator and zygomaticus electromyographic responses) and self-report (valence, arousal ratings) responses were monitored during imagery of anger, fear, joy, and neutral scenes and emotionally neutral high arousal (action) scenes. Results from regression analyses indicated that alexithymia was largely unrelated to responses on valence-based measures (facial electromyography, valence ratings), but that it was related to arousal-based measures. Specifically, alexithymia was related to higher heart rate during neutral and lower heart rate during fear imagery. Alexithymia did not predict differential responses to action versus neutral imagery, suggesting specificity of deficits to emotional contexts. Evidence for discordance between physiological responses and self-report in alexithymia was obtained from within-person analyses using multilevel modeling. Results are consistent with the idea that alexithymic deficits are specific to processing emotional arousal, and suggest difficulties with parasympathetic control and emotion regulation. Alexithymia is also associated with discordance between self-reported emotional experience and physiological response to emotion, consistent with prior evidence.
The English WordNet and the Hindi WordNet have inspired the Gujarati WordNet; this paper specifies the overview and the basic methods for developing a Gujarati WordNet.
See http://clld.org