Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
18265 papers
We argue that coherence relations (relations between propositions, such as Concession or Purpose) are signalled more frequently and by more means than is generally believed. We examine how coherence relations in text are indicated by all possible textual signals, and whether every relation is signalled. To that end, we conducted a corpus study on the RST Discourse Treebank, a corpus of newspaper articles annotated for rhetorical (or coherence) relations. Results from our corpus study show that most relations in text (over 90%) are signalled and also that most signalled relations (over 80%) are indicated not only by discourse markers (and, but, if, since), but also by a wide variety of signals other than discourse markers, such as reference, lexical, semantic, syntactic and graphical features. These findings suggest that signalling of coherence relations is much more sophisticated than previously thought.
This case study of a team in an international workplace investigates processes of language socialization in a transient multilingual setting. Using interview and observational data, the analysis shows how social and linguistic norms are negotiated, with the newcomer positioned as a catalyst for changing language practices toward more English, with the ultimate aim of creating a “global mindset” in the organization. Language socialization in a transient multilingual setting is shown to focus on and assign positive value to new linguistic norms that experienced members are socialized into in a process that hinges on new members functioning as tools for management to bring about the desired change. The article shows that while the newcomer is used as a catalyst for increased use of English and for the creation of a “global mindset,” she is at the same time socialized into the existing Danish egalitarian workplace culture. The analysis also highlights the vulnerable position of the catalyst. While her status as an expert English speaker with the right “global mindset” may confer power in an internationalizing workplace, the position of catalyst is also linked with feelings of exclusion.
Several recent studies have demonstrated that people intuitively make consistent matches between classical music and specific wines. It is not clear, however, what governs such crossmodal mappings. Here, we assess the role of emotion—specifically different dimensional aspects of valence, arousal, and dominance—in mediating such mappings. Participants matched three different red wines to three different pieces of classical music. Subsequently, they made emotion ratings separately for each wine and each musical selection. The results revealed that certain wine–music pairings were rated as being significantly better matches than others. More importantly, there was evidence that the participants’ dominance and arousal ratings for the wines and the music predicted their matching rating for each wine–music pairing. These results therefore support the view that wine–music associations are not arbitrary but can be explained, at least in part, by common emotional associations.
We present a new, efficient frame-semantic parser that labels semantic arguments to FrameNet predicates. Built using an extension to the segmental RNN that emphasizes recall, our basic system achieves competitive performance without any calls to a syntactic parser. We then introduce a method that uses phrase-syntactic annotations from the Penn Treebank during training only, through a multitask objective; no parsing is required at training or test time. This "syntactic scaffold" offers a cheaper alternative to traditional syntactic pipelining, and achieves state-of-the-art performance.
Sentiment analysis on Chinese text has intensively studied. The basic task for related research is to construct an affective lexicon and thereby predict emotional scores of different levels. However, finite lexicon resources make it difficult to effectively and automatically distinguish between various types of sentiment information in Chinese texts. This IJCNLP2017-Task2 competition seeks to automatically calculate Valence and Arousal ratings within the hierarchies of vocabulary and phrases in Chinese. We introduce a regression methodology to automatically recognize continuous emotional values, and incorporate a word embedding technique. In our system, the MAE predictive values of Valence and Arousal were 0.811 and 0.996, respectively, for the sentiment dimension prediction of words in Chinese. In phrase prediction, the corresponding results were 0.822 and 0.489, ranking sixth among all teams.
This paper discusses the use of ‘by’-phrases in Impersonal Passives in Icelandic. It has been claimed in the literature on Icelandic syntax that ‘by’-phrases that express the agent are not very good or even ungrammatical in Impersonal Passives. The pa-per shows that this point of view oversimplifies the facts because various examples of this pattern can be found in natural data and these examples do not seem to reflect mistakes in linguistic performance. We discuss examples from the Icelandic treebank and from the web and we suggest that ‘by’-phrases are more likely to be used in Impersonal Passives if they involve new information and/or if they are heavy. One of the conclusions of the article is that large and well annotated corpora are important for linguistic research that focuses on rare constructions.
Virtual reality (VR) has been proposed as a methodological tool to study the basic science of psychology and other fields. One key advantage of VR is that sharing of virtual content can lead to more robust replication and representative sampling. A database of standardized content will help fulfill this vision. There are two objectives to this study. First, we seek to establish and allow public access to a database of immersive VR video clips that can act as a potential resource for studies on emotion induction using virtual reality. Second, given the large sample size of participants needed to get reliable valence and arousal ratings for our video, we were able to explore the possible links between the head movements of the observer and the emotions he or she feels while viewing immersive VR. To accomplish our goals, we sourced for and tested 73 immersive VR clips which participants rated on valence and arousal dimensions using self-assessment manikins. We also tracked participants' rotational head movements as they watched the clips, allowing us to correlate head movements and affect. Based on past research, we predicted relationships between the standard deviation of head yaw and valence and arousal ratings. Results showed that the stimuli varied reasonably well along the dimensions of valence and arousal, with a slight underrepresentation of clips that are of negative valence and highly arousing. The standard deviation of yaw positively correlated with valence, while a significant positive relationship was found between head pitch and arousal. The immersive VR clips tested are available online as supplemental material.
We present Contrast-Ita Bank, a corpus annotated with discourse contrast relations in Italian. We annotate both explicit and implicit contrast relations, following the schema proposed in the Penn Discourse Treebank. We provide and discuss quantitative data about the new resource.
Previous research has shown that good looks, particularly being deemed as attractive or competent-looking, can provide an electoral advantage. There is also evidence to support the notion that more dominant looks are associated with military success as cadets are more likely to rise in the ranks early in their career if they are more dominant-looking. To date, there has been little research into the effect of looks on political leadership success in a non democratic setting. This project explores the effect of facial attractiveness and dominance on the political success of leaders after leading a successful coup d’état. We examine a comprehensive set of coup d'états from 1946 to 2013. Attractiveness and dominance ratings are created via surveys, as in previous research, but with a novel way to control for the potential bias arising from respondent characteristics. Defining political success as taking executive power, longer time-to-office exit, and avoiding constraint on executive power, we find that both dominant and attractive facial features provide distinct advantages for leaders.
The article describes the process of retaining the recessive units in the correctness publications which have been the sources of the codified norm for the last hundred years. This includes the following forms: interesa, czochrze, w Prusiech. Such elements marked i.a. as rare, former, out of use or outdated belonged in the given period to the linguistic norm of some speakers of Polish. Therefore linguists made attempts to retain them in the codified norm at least for some time so that they could serve as evidence of their former correctness. They were used with qualifiers which provided information about potential question concerning the topicality of the recessive elements. Such units met different fates. They were no longer provided or perceived as wrong in the succeeding dictionaries that were examined. They often functioned as recessive until contemporaneity or even returned as equal variants. A detailed typology of these relations was discussed in the article.
This paper presents a novel neural machine translation model which jointly learns translation and source-side latent graph representations of sentences. Unlike existing pipelined approaches using syntactic parsers, our end-to-end model learns a latent graph parser as part of the encoder of an attention-based neural machine translation model, and thus the parser is optimized according to the translation objective. In experiments, we first show that our model compares favorably with state-of-the-art sequential and pipelined syntax-based NMT models. We also show that the performance of our model can be further improved by pretraining it with a small amount of treebank annotations. Our final ensemble model significantly outperforms the previous best models on the standard Englishto-Japanese translation dataset.
<h3>Introduction</h3><br> Abstract Meaning Representation (AMR) Annotation Release 2.0 was developed by the Linguistic Data Consortium (LDC), <a href="http://www.sdl.com/">SDL/Language Weaver, Inc.</a>, the University of Colorado's <a href="http://clear.colorado.edu/start/index.html">Computational Language and Educational Research</a> group and the <a href="http://www.isi.edu/home">Information Sciences Institute</a> at the University of Southern California. It contains a sembank (semantic treebank) of over 39,260 English natural language sentences from broadcast conversations, newswire, weblogs and web discussion forums. <br> AMR captures “who is doing what to whom” in a sentence. Each sentence is paired with a graph that represents its whole-sentence meaning in a tree-structure. AMR utilizes PropBank frames, non-core semantic roles, within-sentence coreference, named entity annotation, modality, negation, questions, quantities, and so on to represent the semantic structure of a sentence largely independent of its syntax. <br> LDC also released Abstract Meaning Representation (AMR) Annotation Release 1.0 (<a href="../../../LDC2014T12">LDC2014T12</a>). <br> <h3>Data</h3><br> The source data includes discussion forums collected for the DARPA BOLT and DEFT programs, transcripts and English translations of Mandarin Chinese broadcast news programming from China Central TV, Wall Street Journal text, translated Xinhua news texts, various newswire data from NIST OpenMT evaluations and weblog data used in the DARPA GALE program. The following table summarizes the number of training, dev, and test AMRs for each dataset in the release. Totals are also provided by partition and dataset: <br> <table border="1" cellpadding="2"><br> <tbody><br> <tr><br> <td>Dataset</td><br> <td>Training</td><br> <td>Dev</td><br> <td>Test</td><br> <td>Totals</td><br> </tr><br> <tr><br> <td>BOLT DF MT</td><br> <td>1061</td><br> <td>133</td><br> <td>133</td><br> <td>1327</td><br> </tr><br> <tr><br> <td>Broadcast conversation</td><br> <td>214</td><br> <td>0</td><br> <td>0</td><br> <td>214</td><br> </tr><br> <tr><br> <td>Weblog and WSJ</td><br> <td>0</td><br> <td>100</td><br> <td>100</td><br> <td>200</td><br> </tr><br> <tr><br> <td>BOLT DF English</td><br> <td>6455</td><br> <td>210</td><br> <td>229</td><br> <td>6894</td><br> </tr><br> <tr><br> <td>DEFT DF English</td><br> <td>19558</td><br> <td>0</td><br> <td>0</td><br> <td>19558</td><br> </tr><br> <tr><br> <td>Guidelines AMRs</td><br> <td>819</td><br> <td>0</td><br> <td>0</td><br> <td>819</td><br> </tr><br> <tr><br> <td>2009 Open MT</td><br> <td>204</td><br> <td>0</td><br> <td>0</td><br> <td>204</td><br> </tr><br> <tr><br> <td>Proxy reports</td><br> <td>6603</td><br> <td>826</td><br> <td>823</td><br> <td>8252</td><br> </tr><br> <tr><br> <td>Weblog</td><br> <td>866</td><br> <td>0</td><br> <td>0</td><br> <td>866</td><br> </tr><br> <tr><br> <td>Xinhua MT</td><br> <td>741</td><br> <td>99</td><br> <td>86</td><br> <td>926</td><br> </tr><br> <tr><br> <td>Totals</td><br> <td>36521</td><br> <td>1368</td><br> <td>1371</td><br> <td>39260</td><br> </tr><br> </tbody><br> </table><br> <br> For those interested in utilizing a standard/community partition for AMR research (for instance in development of semantic parsers), data in the "split" directory contains 39,260 AMRs split roughly 93%/3.5%/3.5% into training/dev/test partitions, with most smaller datasets assigned to one of the splits as a whole. Note that splits observe document boundaries. The "unsplit" directory contains the same 39,260 AMRs with no train/dev/test partition. <br> <h3>Samples</h3><br> Please view this <a href="desc/addenda/LDC2017T10.xml">sample</a>. <br> <h3>Updates</h3><br> None at this time. <br> <h3>Acknowledgements</h3><br> From University of Colorado <br> We gratefully acknowledge the support of the National Science Foundation Grant NSF: 0910992 IIS:RI: Large: Collaborative Research: Richer Representations for Machine Translation and the support of Darpa BOLT - HR0011-11-C-0145 and DEFT - FA-8750-13-2-0045 via a subcontract from LDC. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation, DARPA or the US government. <br> From Information Sciences Institute (ISI) <br> Thanks to NSF (IIS-0908532) for funding the initial design of AMR, and to DARPA MRP (FA-8750-09-C-0179) for supporting a group to construct consensus annotations and the AMR Editor. The initial AMR bank was built under DARPA DEFT FA-8750-13-2-0045 (PI: Stephanie Strassel; co-PIs: Kevin Knight, Daniel Marcu, and Martha Palmer) and DARPA BOLT HR0011-12-C-0014 (PI: Kevin Knight). <br> From Linguistic Data Consortium (LDC) <br> This material is based on research sponsored by Air Force Research Laboratory and Defense Advance Research Projects Agency under agreement number FA8750-13-2-0045. The U.S. Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright notation thereon. The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of Air Force Research Laboratory and Defense Advanced Research Projects Agency or the U.S. Government. <br> We gratefully acknowledge the support of Defense Advanced Research Projects Agency (DARPA) Machine Reading Program under Air Force Research Laboratory (AFRL) prime contract no. FA8750-09-C-0184 Subcontract 4400165821. Any opinions, findings, and conclusion or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the view of the DARPA, AFRL, or the US government. <br> From Language Weaver (SDL) <br> This work was partially sponsored by DARPA contract HR0011-11-C-0150 to LanguageWeaver Inc. Any opinions, findings, and conclusion or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the view of the DARPA or the US government. </br> Portions © 2002-2005, 2007-2008 Agence France Presse, © 2007 Al Ahram, © 2007 Al Hayat, © 2007 Al-Quds Al-Arabi, © 2007 Asharq Al-Awsat, © 2007 An Nahar, © 2007 Assabah, © 2002-2008 The Associated Press, © 2003-2004, 2007-2008 Central News Agency (Taiwan), © 1997, 2004-2007 China Central TV, © 2007 China Military Online, © 2007 Chinanews.com, © 1987-1989 Dow Jones & Company, Inc., © 2007 Guangming Daily, © 1995, 2003, 2007-2008 Los Angeles Times-Washington Post News Service, Inc., © 2002, 2004-2005, 2007-2008 New York Times, © 1994-1998, 2001-2008 Xinhua News Agency, © 2014, 2017 Language Weaver, Inc., © 2014, 2017 University of Colorado, © 2014, 2017 University of Southern California, © 2003, 2005, 2006, 2007, 2009, 2011, 2013, 2014, 2017 Trustees of the University of Pennsylvania
This article presents the issue of linguistic policy and the attitude of educational authorities towards dialects from the point of view of the multi-lingual and multiculture Italian society. Due to specific historical conditions the linguistic situation in Italy is characterized by a high degree of heterogeneity -numerous dialects exist alongside the national language. After the period of the political unity of Italy in 1861 the educational authorities have had to tackle two problems: the necessity of having to choose a linguistic norm and also mutual relations between the language and the dialects. Those issues were and still are highly controversial and are subject to heated polemics as the answer to the question of whether to accept dialects or eliminate them from school education has never been found.
Word embeddings that can capture semantic and syntactic information from contexts have been extensively used for various natural language processing tasks. However, existing methods for learning contextbased word embeddings typically fail to capture sufficient sentiment information. This may result in words with similar vector representations having an opposite sentiment polarity (e.g., good and bad), thus degrading sentiment analysis performance. Therefore, this study proposes a word vector refinement model that can be applied to any pre-trained word vectors (e.g., Word2vec and GloVe). The refinement model is based on adjusting the vector representations of words such that they can be closer to both semantically and sentimentally similar words and further away from sentimentally dissimilar words. Experimental results show that the proposed method can improve conventional word embeddings and outperform previously proposed sentiment embeddings for both binary and fine-grained classification on Stanford Sentiment Treebank (SST).
The Web has accumulated a rich source of information, such as text, image, rating, etc, which represent different aspects of user preferences. However, the heterogeneous nature of this information makes it difficult for recommender systems to leverage in a unified framework to boost the performance. Recently, the rapid development of representation learning techniques provides an approach to this problem. By translating the various information sources into a unified representation space, it becomes possible to integrate heterogeneous information for informed recommendation.
Objective: Abnormalities in emotion processing are increasingly discussed in patients with anorexia nervosa (AN). In addition, food avoidance and attentional biases in the context of food are important issues in the investigation of eating disorder pathology. The aim of this study was to investigate visual attention in reaction to emotional pictures and food stimuli with eye-tracking technology. Furthermore, it should be investigated, if there were differences between adolescent and adult patients associated with the chronification of the disorder. Changes of the effects during the inpatient treatment should also be considered. Method: 41 female AN patients and 38 controls (43 adults and 36 adolescents) were presented with 50 disorder specific food pictures and emotional stimuli (neutral, positive, negative). Their task was to evaluate valence and arousal of each stimulus. An eye-tracking camera registered fixation times within and outside of specific regions of interest. In another experiment, they were presented with 20 pairs of a high caloric and a low caloric stimulus each. Intelligence, eating disorder pathology and general psychopathology were assessed by tests and questionnaires. AN patients performed the experiment twice (i.e. before and after inpatient treatment). Results: Valence and arousal ratings for the emotional pictures were comparable for the group of AN patients and controls. High caloric stimuli were rated more negative in valence and with higher arousal by AN patients compared to controls. Between the age groups significant differences concerning ratings of picture categories neutral, positive and low caloric were found which were not disorder specific. Valence ratings of categories high and low caloric changed significantly during the course of treatment toward the neutral value. All participants fixated longer within the ROIs than within the rest of the pictures. The relative fixation times of the controls for the low and high caloric stimuli did not differ. In contrast, among the patients, longer fixations were measured for the high caloric stimuli. In total, low caloric stimuli were fixated significantly longer than high caloric stimuli in the ”pairs” experiment. The AN patients fixated the low caloric stimuli longer compared to controls. The lower the BMI, the longer the participants fixated the ROIs of low caloric pictures. Also regarding eye tracking measures, there were no differences between adolescent and adult patients. After the inpatient treatment, the difference in fixation duration between the high and low caloric stimulus became smaller. Conclusions: AN patiens showed an attional bias towards disorder specific stimuli (food pictures). If the patiens direct their attention more or less towards high caloric stimuli seems to depend on the context. Altogether, the results may indicate aversion against high caloric food stimuli, while the effects were stronger for the ratings than for the eye-tracking. There were not found any indications for a deviation in emotion regulation. Although some differences were observed between adolescents and adults, the result pattern for adolescent and adult patients was similar and, thus, not to be related to the chronification of the disorder. Because of the non-controlled setting, it can´t be differentiated if the changes after the inpatient treatment represent therapeutic effects or whether they reflect repetition of the test. Even though further research is required the findings of the study may contribute to a better understanding of anorexia nervosa.
Discourse parsing is an integral part of understanding information flow and argumentative structure in documents. Most previous research has focused on inducing and evaluating models from the English RST Discourse Treebank. However, discourse treebanks for other languages exist, including Spanish, German, Basque, Dutch and Brazilian Portuguese. The treebanks share the same underlying linguistic theory, but differ slightly in the way documents are annotated. In this paper, we present (a) a new discourse parser which is simpler, yet competitive (significantly better on 2/3 metrics) to state of the art for English, (b) a harmonization of discourse treebanks across languages, enabling us to present (c) what to the best of our knowledge are the first experiments on crosslingual discourse parsing.
This paper presents the first version of a Danish Propbank/VerbNet corpus, annotated at both the morphosyntactic, dependency and semantic levels. Both verbal and nominal predications were tagged with frames consisting of a VerbNet class and semantic role-labeled arguments and satellites. As a second semantic annotation layer, the corpus was tagged with both a noun ontology and NER classes. Drawing on mixed news, magazine, blog and forum data from DSL's Korpus2010, the 87,000 token corpus contains over 12,000 frames with 32,000 semantic role instances. We discuss both technical and linguistic aspects of the annotation process, evaluate coverage and provide a statistical break-down of frames and roles for both the corpus as a whole and across different text types.
Syntactic discontinuities are very frequent in classical Latin and yet this data was never considered in debates on how expressive grammar formalisms need to be to capture natural languages. In this paper I show with treebank data that Latin frequently displays syntactic discontinuities that cannot be captured in standard mildly context-sensitive frameworks such as Tree-Adjoining Grammars or Combinatory Categorial Grammars. I then argue that there is no principled bound on Latin discontinuities but that they display a broadly Zipfian distribution where frequency drops quickly for the more complex patterns. Lexical-Functional Grammar can capture these discontinuities in a way that closely reflects their complexity and frequency distributions.
Recent papers have shown that neural networks obtain state-of-the-art performance on several different sequence tagging tasks. One appealing property of such systems is their generality, as excellent performance can be achieved with a unified architecture and without task-specific feature engineering. However, it is unclear if such systems can be used for tasks without large amounts of training data. In this paper we explore the problem of transfer learning for neural sequence taggers, where a source task with plentiful annotations (e.g., POS tagging on Penn Treebank) is used to improve performance on a target task with fewer available annotations (e.g., POS tagging for microblogs). We examine the effects of transfer learning for deep hierarchical recurrent networks across domains, applications, and languages, and show that significant improvement can often be obtained. These improvements lead to improvements over the current state-of-the-art on several well-studied tasks.
Predicting emotion intensity and severity of depression are both challenging and important problems within the broader field of affective computing. As part of the AVEC 2017, we developed a number of systems to accomplish these tasks. In particular, word affect features, which derive human affect ratings (e.g. arousal and valence) from transcripts, were investigated for predicting depression severity and liking, showing great promise. A simple system based on the word affect features achieved an RMSE of 6.02 on the test set, yielding a relative improvement of 13.6% over the baseline. For the emotion prediction sub-challenge, we investigated multimodal fusion, which incorporated a measure of uncertainty associated with each prediction within an Output-Associative fusion framework for arousal and valence prediction, whilst liking prediction systems mainly focused on text-based features. Our best emotion prediction systems provided significant relative improvements over the baseline on the test set of 39.5%, 17.6%, and 29.3% for arousal, valence, and liking. Of particular note is that consistent improvements were observed when incorporating prediction uncertainty across various system configurations for predicting arousal and valence, suggesting the importance of taking into consideration prediction uncertainty for fusion and more broadly the advantages of probabilistic predictions.
In this paper, a deep phrase embedding approach using bi-directional long short-term memory (Bi-LSTM) is proposed to predict the valence-arousal ratings of Chinese words and phrases. It adopts a Chinese word segmentation frontend, a local order-aware word, a global phrase embedding representations and a deep regression neural network (DRNN) model. The performance of the proposed method was benchmarked by the IJCNLP 2017 shared task 2. According the official evaluation results, our best system achieved mean rank 6.5 among all 24 submissions.
The isolation of errors in the language of young children has led to general and interesting statements concerning the acquisition process. This chapter examines the notion of error itself from a sociolinguistic perspective. It suggests that the notion of error should be understood in light of the range of variants in use for the feature under consideration. The chapter demonstrates that the assessment of what counts as an error is affected by data collection methodology, in particular by the way in which adult linguistic norms are established. If formality of speech can be measured in terms of degree of attention to one's speech, then these circumstances of language use are relatively formal. The chapter relates these theoretical and methodological points to research carried out on Samoan language acquisition, and then focuses on selected features of children's language.
ForFun is a database of linguistic forms and their syntactic functions built with the use of the multi-layer annotated corpora of Czech, the Prague Dependency Treebanks. The purpose of the Prague Database of Forms and Functions (ForFun) is to help the linguists to study the form-function relation, which we assume to be one of the principal tasks of both theoretical linguistics and natural language processing. A prototypical question to be asked is What purposes does a preposition 'po' serve for or What are the linguistic means in the sentence that can express the meaning 'a destination of an action'?. There are almost 1500 distinct forms (besides the 'po' preposition) and 65 distinct functions (besides the 'destination').
Dependency parsing is considered as the state of the art technology for a better information extraction methodology in Natural Language Processing. With the ever-growing need for linguistic analysis for different languages, the demand for multilingual dependency parsing has increased dramatically. In this research work we studied a novel collection of treebanks with homogeneous syntactic dependency annotation [1] for six languages along with other recent techniques in this area. We investigated the possibility of adding new languages in this module and successfully added universal Bangla dependency annotation. Additionally, we combined simple and complex feature representations to improve parsing output.
This chapter provides an overview of language policy and the media by reviewing the state of the art, both in terms of literature and in terms of research. It outlines key terms and their uses, and explains the types of language policy and media. The chapter also provides an overview of disciplinary perspectives on language policy and the media, with a particular focus on the evolution from traditional news to new and social media. It reviews research within a range of national and globalized contexts, and discusses the core areas of status, corpus and acquisition planning. The chapter examines relationship between language policy and the media in two overarching areas: in the chronological transition from nation-states to globalization, and in status, corpus, and acquisition planning. Language policy concerns the production and enforcement of linguistic norms; the reality of policymaking and its implementation is much more complex than such a simplistic label implies.
With the progress of science, we observe an unprecedented use of adverbial participles, which come to express increasingly more complex concepts and relations. This tendency is evident not only in the texts of natives, but also of non-natives. The present study examines to what extent non-native speakers of English are influenced by linguistic norms of their native languages when writing academic texts in English. It also focuses on the role of adverbial participles in the syntactic and informative organization of scientific English. The quantitative-qualitative analysis of the material has revealed that the encoding and transmission of complex ideas in scientific English require a high degree of coherence. The results of the paper are sure to contribute to current research in applied and corpus linguistics from the perspective of speakers’ cognitive processes and their linguistic realizations.
This paper formalizes a sound extension of dynamic oracles to global training, in the frame of transition-based dependency parsers. By dispensing with the precomputation of references, this extension widens the training strategies that can be entertained for such parsers; we show this by revisiting two standard training procedures, early-update and max-violation, to correct some of their search space sampling biases. Experimentally, on the SPMRL treebanks, this improvement increases the similarity between the train and test distributions and yields performance improvements up to 0.7 UAS, without any computation overhead.
This paper reports on a suite of experiments that evaluates how the linguistic granularity of part-of-speech tagsets impacts the performance of tagging and syntactic dependency parsing. Our results show that parsing accuracy can be significantly improved by introducing more finegrained morphological information in the tagset, even if tagger accuracy is compromised. Our taggers and parsers are trained and tested using the annotations of the Norwegian Dependency Treebank.
Long short-term memory (LSTM) has been widely used in different applications, such as natural language processing, speech recognition, and computer vision over recurrent neural network (RNN) or recursive neural network (RvNN)-a tree-structured RNN. In addition, the LSTM-RvNN has been used to represent compositional semantics through the connections of hidden vectors over child units. However, the linear connections in the existing LSTM networks are incapable of capturing complex semantic representations of natural language texts. For example, complex structures in natural language texts usually denote intricate relationships between words, such as negated sentiment or sentiment strengths. In this paper, quadratic connections of the LSTM model is proposed in terms of RvNNs (abbreviated as qLSTM-RvNN) in order to attack the problem of representing compositional semantics. The proposed qLSTM-RvNN model is evaluated in the benchmark data sets containing semantic compositionality, i.e., sentiment analysis on Stanford Sentiment Treebank and semantic relatedness on sentences involving compositional knowledge data set. Empirical results show that it outperforms the state-of-the-art RNN, RvNN, and LSTM networks in two semantic compositionality tasks by increasing the classification accuracies and sentence correlation while significantly decreasing computational complexities.
Truncated Backpropagation Through Time (truncated BPTT) is a widespread method for learning recurrent computational graphs. Truncated BPTT keeps the computational benefits of Backpropagation Through Time (BPTT) while relieving the need for a complete backtrack through the whole data sequence at every step. However, truncation favors short-term dependencies: the gradient estimate of truncated BPTT is biased, so that it does not benefit from the convergence guarantees from stochastic gradient theory. We introduce Anticipated Reweighted Truncated Backpropagation (ARTBP), an algorithm that keeps the computational benefits of truncated BPTT, while providing unbiasedness. ARTBP works by using variable truncation lengths together with carefully chosen compensation factors in the backpropagation equation. We check the viability of ARTBP on two tasks. First, a simple synthetic task where careful balancing of temporal dependencies at different scales is needed: truncated BPTT displays unreliable performance, and in worst case scenarios, divergence, while ARTBP converges reliably. Second, on Penn Treebank character-level language modelling, ARTBP slightly outperforms truncated BPTT.
Causal relations play a key role in information extraction and reasoning. Most of the times, their expression is ambiguous or implicit, i.e. without signals in the text. This makes their identification challenging. We aim to improve their identification by implementing a Feedforward Neural Network with a novel set of features for this task. In particular, these are based on the position of event mentions and the semantics of events and participants. The resulting classifier outperforms strong baselines on two datasets (the Penn Discourse Treebank and the CSTNews corpus) annotated with different schemes and containing examples in two languages, English and Portuguese. This result demonstrates the importance of events for identifying discourse relations.
The appreciation of humor is a universal phenomenon and a key aspect of cognition. It has been studied in thecontext of jokes, where the incongruity in expected and observed context results in the perception of humor. The present studyexamines how the humor appreciation of single words relates to the humor of the whole joke – is a joke simply a sum of itsparts? Using a novel dataset of single-word humor ratings, collections of jokes from the JESTER database were analyzed. Amultiple regression analysis showed joke length and individual word arousal were the best predictors of joke funniness. Longerjokes with fewer individually arousing words were found funnier. Individual word humor did not contribute to the humor ofthe overall joke. These findings suggest the cognitive aspects of humor are likely driven by broader semantic context, whereasappreciating humor on a per-word basis links to separate factors.
Part-of-speech (POS) tagging for morphologically rich languages such as Arabic is a challenging problem because of their enormous tag sets. One reason for this is that in the tagging scheme for such languages, a complete POS tag is formed by combining tags from multiple tag sets defined for each morphosyntactic category. Previous approaches in Arabic POS tagging applied one model for each morphosyntactic tagging task, without utilizing shared information between the tasks. In this paper, we propose an approach that utilizes this information by jointly modeling multiple morphosyntactic tagging tasks with a multi-task learning framework. We also propose a method of incorporating tag dictionary information into our neural models by combining word representations with representations of the sets of possible tags. Our experiments showed that the joint model with tag dictionary information results in an accuracy of 91.38% on the Penn Arabic Treebank data set, with an absolute improvement of 2.11% over the current state-of-the-art tagger. 1
We present UCCAApp, an open-source, flexible web-application for syntactic and semantic phrase-based annotation in general, and for UCCA annotation in particular.UCCAApp supports a variety of formal properties that have proven useful for syntactic and semantic representation, such as discontiguous phrases, multiple parents and empty elements, making it useful to a variety of other annotation schemes with similar formal properties.UCCAApp's user interface is intuitive and user friendly, so as to support annotation by users with no background in linguistics or formal representation.Indeed, a pilot version of the application has been successfully used in the compilation of the UCCA Wikipedia treebank by annotators with no previous linguistic training.The application and all accompanying resources are released as open source under the GNU public license, and are available online along with a live demo. 1
Abstract In the paper, we present our efforts to annotate evaluative language in the Prague Dependency Treebank 2.0. The project is a follow-up of the series of annotations of small plaintext corpora. It uses automatic identification of potentially evaluative nodes through mapping a Czech subjectivity lexicon to syntactically annotated data. These nodes are then manually checked by an annotator and either dismissed as standing in a non-evaluative context, or confirmed as evaluative. In the latter case, information about the polarity orientation, the source and target of evaluation is added by the annotator. The annotations unveiled several advantages and disadvantages of the chosen framework. The advantages involve more structured and easy-to-handle environment for the annotator, visibility of syntactic patterning of the evaluative state, effective solving of discontinuous structures or a new perspective on the influence of good/bad news. The disadvantages include little capability of treating cases with evaluation spread among more syntactically connected nodes at once, little capability of treating metaphorical expressions, or disregarding the effects of negation and intensification in the current scheme.
We describe a simple but effective method for cross-lingual syntactic transfer of dependency parsers, in the scenario where a large amount of translation data is not available. This method makes use of three steps: 1) a method for deriving cross-lingual word clusters, which can then be used in a multilingual parser; 2) a method for transferring lexical information from a target language to source language treebanks; 3) a method for integrating these steps with the density-driven annotation projection method of Rasooli and Collins (2015). Experiments show improvements over the state-of-the-art in several languages used in previous work, in a setting where the only source of translation data is the Bible, a considerably smaller corpus than the Europarl corpus used in previous work. Results using the Europarl corpus as a source of translation data show additional improvements over the results of Rasooli and Collins (2015). We conclude with results on 38 datasets from the Universal Dependencies corpora.
We present a novel transition system, based on the Covington non-projective parser, introducing non-local transitions that can directly create arcs involving nodes to the left of the current focus positions. This avoids the need for long sequences of No-Arc transitions to create long-distance arcs, thus alleviating error propagation. The resulting parser outperforms the original version and achieves the best accuracy on the Stanford Dependencies conversion of the Penn Treebank among greedy transition-based algorithms.
Theoretical models and recent advances in the treatment of anorexia nervosa (AN) have increasingly focused on the role of alterations in the processing and regulation of emotions. To date, however, our understanding of these changes is still limited and reports of emotional dysregulation in AN have been based largely on self-report data, and there is a relative lack of objective experimental evidence or neurobiological data. The current functional magnetic resonance imaging (fMRI) study investigated the hemodynamic correlates of passive viewing and voluntary downregulation of negative emotions by means of the reappraisal strategy detachment in AN patients. Detachment is regarded as adaptive regulation strategy associated with a reduction in emotion-related amygdala activity and increased recruitment of prefrontal brain regions associated with cognitive control processes. Emotion regulation efficacy was assessed via behavioral arousal ratings and fMRI activation elicited by an established experimental paradigm including negative images. Participants were instructed to either simply view emotional pictures or detach themselves from feelings triggered by the stimuli. The sample consisted of 36 predominantly adolescent female AN patients and a pairwise age-matched healthy control group. Behavioral and neuroimaging data analyses indicated a reduction of arousal and amygdala activity during the regulation condition for both patients and controls. However, compared with controls, individuals with AN showed increased activation in the amygdala as well as in the right dorsolateral prefrontal cortex (dlPFC) during the passive viewing of aversive compared with neutral pictures. These results extend previous findings indicative of altered processing of salient emotional stimuli in AN, but do not point to a general deficit in the voluntary regulation of negative emotions. Increased dlPFC activation in AN during passive viewing of negative stimuli is in line with the hypothesis that the disorder may be characterized by excessive self-control. Taken together, the data seem to suggest that reappraisal via detachment may be an effective strategy to reduce negative arousal for individuals with AN.
This paper presents our submissions for the CoNLL 2017 UD Shared Task. Our parser, called UParse, is based on a neural network graph-based dependency parser. The parser uses features from a bidirectional LSTM to produce a distribution over possible heads for each word in the sentence. To allow transfer learning for lowresource treebanks and surprise languages, we train several multilingual models for related languages, grouped by their genus and language families. Out of 33 participants, our system achieves rank 9th in the main results, with 75.49 UAS and 68.87 LAS F-1 scores (average across 81 treebanks).
The LyS-FASTPARSE team presents BIST-COVINGTON, a neural implementation of the Covington ( The bidirectional LSTM approach by Kiperwasser and Goldberg ( The model participated in the CoNLL 2017 UD Shared Task. In spite of not using any ensemble methods and using the baseline segmentation and PoS tagging, the parser obtained good results on both macro-average LAS and UAS in the big treebanks category (55 languages), ranking 7th out of 33 teams. In the all treebanks category (LAS and UAS) we ranked 16th and 12th. The gap between the all and big categories is mainly due to the poor performance on four parallel PUD treebanks, suggesting that some 'suffixed' treebanks (e.g. Spanish-AnCora) perform poorly on cross-treebank settings, which does not occur with the corresponding 'unsuffixed' treebank (e.g. Spanish). By changing that, we obtain the 11th best LAS among all runs (official and unofficial). The code is made available at https://github.com/CoNLL-UD-2017/LyS-FASTPARSE
General treebank analyses are graph structured, but parsers are typically restricted to tree structures for efficiency and modeling reasons. We propose a new representation and algorithm for a class of graph structures that is flexible enough to cover almost all treebank structures, while still admitting efficient learning and inference. In particular, we consider directed, acyclic, one-endpoint-crossing graph structures, which cover most long-distance dislocation, shared argumentation, and similar tree-violating linguistic phenomena. We describe how to convert phrase structure parses, including traces, to our new representation in a reversible manner. Our dynamic program uniquely decomposes structures, is sound and complete, and covers 97.3% of the Penn English Treebank. We also implement a proof-of-concept parser that recovers a range of null elements and trace types.
First paragraph: The Scots language has largely been excluded, historically, within Scottish institutional contexts (Jones 1995: 1-21). This phenomenon typically owes itself, in Bourdieuan terms, to the lack of 'social' and 'cultural capital' certain codes of the language have increasingly held since the eighteenth century onwards in much of Scottish society. The devaluation of the Scots language from this period has been exacerbated in particular by its growing marginalisation within the Scottish education system. Although learning Latin held prestige during the sixteenth and seventeenth centuries, Scots was generally the teaching medium in most Scottish classrooms (Williamson 1982a: 54-77). However the elocution movement during the latter half of the eighteenth century and the Education (Scotland) Act of 1872, both encouraged and eventually required that every child should be educated in English (Bailey 1987: 131-42). Scots became regarded as a 'lazy', parochial dialect of English and Scottish aspirations to reproduce the linguistic norms of 'polite' London helped to suppress the language further (Jones 1995: 2).
Increased pornography use has been a feature of contemporary human society, with technological advances allowing for high speed internet and relative ease of access via a multitude of wireless devices. Does increased pornography exposure alter general emotion processing? Research in the area of pornography use is heavily reliant on conscious self-report measures. However, increasing knowledge indicates that attitudes and emotions are extensively processed on a non-conscious level prior to conscious appraisal. Hence, this exploratory study aimed to investigate whether frequency of pornography use has an impact on non-conscious and/or conscious emotion processes. Participants (N = 52) who reported viewing various amounts of pornography were presented with emotion inducing images. Brain Event-Related Potentials (ERPs) were recorded and Startle Reflex Modulation (SRM) was applied to determine non-conscious emotion processes. Explicit valence and arousal ratings for each image presented were also taken to determine conscious emotion effects. Conscious explicit ratings revealed significant differences with respect to “Erotic” and “Pleasant” valence (pleasantness) ratings depending on pornography use. SRM showed effects approaching significance and ERPs showed changes in frontal and parietal regions of the brain in relation to “Unpleasant” and “Violent” emotion picture categories, which did not correlate with differences seen in the explicit ratings. Findings suggest that increased pornography use appears to have an influence on the brain’s non-conscious responses to emotion-inducing stimuli which was not shown by explicit self-report.
Most of the machine learning algorithms requires the input to be denoted as a fixed-length feature vector. In text classifications (bag-of-words) is a popular fixed-length features. Despite their simplicity, they are limited in many tasks; they ignore semantics of words and loss ordering of words. In this paper, we propose a simple and efficient neural language model for sentence-level classification task. Our model employs Recurrent Neural Network Language Model (RNN-LM). Particularly, Long Short-Term Memory (LSTM) over pre-trained word vectors obtained from unsupervised neural language model to capture semantics and syntactic information in a short sentence. We achieved outstanding empirical results on multiple benchmark datasets, IMDB Sentiment analysis dataset, and Stanford Sentiment Treebank (SSTb) dataset. The empirical results show that our model is comparable with neural methods and outperforms traditional methods in sentiment analysis task.
To facilitate cross-lingual studies, there is an increasing interest in identifying linguistic universals. Recently, a new universal scheme was designed as a part of universal dependency project. In this paper, we map the Arabic tweets dependency treebank (ATDT) to the Universal Dependency (UD) scheme to compare it to other language resources and for the purpose of cross-lingual studies.
International audience
Relational processing underpins many forms of human thinking. This research addressed whether relational integration in the n-term task (linear syllogisms with three and four premises) can be facilitated. We hypothesised that solving distant analogies (e.g. furnace: coal:: stomach: ______) but not near analogies (e.g. furnace: coal:: woodstove:______) would facilitate n-term performance of undergraduates (N = 120). Participants generated solutions to near analogies (near condition), to distant analogies (distant condition) while participants in the control condition completed a word rating task. Participants then completed the n-term task with items at two levels of complexity (ternary, quaternary), and a fluid intelligence test. Solving distant analogies facilitated relational integration on the more complex quaternary-relational items. It eliminated the complexity effect in the n-term task and the association between quaternary-relational reasoning and fluid intelligence.