Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
16504 papers
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.
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.
Digitization has changed the concept of dictionaries from merely alphabetically ordered reference works into lexical databases providing flexible search systems with interconnected lemmas. This article investigates ensuing opportunities and useful design options of digitized historical dictionaries as research tools for the study of texts. It appears that we have arrived at an interesting intersection of digital humanities and historical lexicography. The 14th-century ‘seemly play of Winter and Summer’ serves as a research case.
Cultural determinants of language behavior and the practice of foreign language teaching in the perspective of socio-cultural normsThe article discusses the problem of the existing variants of socio-cultural behaviour including socio-linguistic behaviour, based on the example of the Polish community in Ukraine. The author considers the possibility of accepting deviations from some of the socio-cultural norms considered in Poland as acceptable variants of the norms, similarly to the linguistic norms. In the second part of the article the author analyses students’ statements considered as incorrect in terms of the possibility of accepting them as alternative forms of socio-cultural norms.
Deaf individuals usually face more challenges in reading and writing, because they are often deprived of adequate spoken input from their infancy. Research on the language features of deaf individuals’ writing is abundant. However, their language structures have as yet been unexplored. In order to address this subject, this article uses the holistic approach of complex network theory. This study builds three syntactic dependency networks, the intent being to capture the macroscopic linguistic features in writing of deaf individuals. Three networks are constructed: one is created from a treebank of texts produced by deaf individuals, and the other two are created from two treebanks of spoken and written language samples produced by hearing people. A dependency‐based theory of syntax is used. The results indicate that the language system of individuals with deafness is structurally similar to that of hearing people, especially to that of their spoken language, but individuals with deafness tend to have lower language proficiency in both syntactic and lexical aspects. The rigid use of function words and less diversity of vocabulary might be part of the reason for the observed differences.
The Estonian Dependency Treebank and its theoretical basis This article presents the Estonian Dependency Treebank (EDT) and discusses its language-theoretical basis. EDT contains ca 400,000 tokens of fiction, newspaper and science texts. Its syntactic annotation is based on principles of dependency syntax. Previous experiments with annotating Estonian sentences according to the principles of phrase structure syntax have shown that the resulting trees tend to be too shallow and thus do not encode the linguistic information in the best possible way. Therefore dependency-syntactic representation was chosen instead. Dependency relations are efficient for encoding typical head-dependent relations like verb-argument or head-modifier but are not so suitable for analysing adpositional phrases, verbal chains, multi-word expressions or other constructs without clear internal syntactic relations. In such cases, there are arguments both for and against all possible solutions.
Slovak models for MorphoDiTa, providing morphological analysis, morphological generation and part-of-speech tagging. The morphological dictionary is created from MorfFlex SK 170914 and the PoS tagger is trained on automatically translated Prague Dependency Treebank 3.0 (PDT).
The problem of (semi-)automatic treebank conversion arises when converting between different schemas, such as from a language specific schema to Universal Dependencies, or when converting from one Universal Dependencies version to the next. This thesis develops a formalism based on top-down tree transducers to convert dependency trees. Building on a well-defined mechanism yields a robust transformation system with clear semantics for rules and which guarantees that every transformation step results in a well formed tree, in contrast to previously proposed solutions. To exemplify the efficiency of the approach, a rule set to transform the Hamburg Dependency Treebank is created, which can already transform annotations with both coverage and precision of more than 90%.
Syntactic parsing is one of the fundamental tasks of Natural Language Processing (NLP). However, few studies have explored syntactic parsing in the medical domain. This dissertation systematically investigated different methods to improve the performance of syntactic parsing of clinical text, including (1) Constructing two clinical treebanks of discharge summaries and progress notes by developing annotation guidelines that handle missing elements in clinical sentences; (2) Retraining four state-of-the-art parsers, including the Stanford parser, Berkeley parser, Charniak parser, and Bikel parser, using clinical treebanks, and comparing their performance to identify better parsing approaches; and (3) Developing new methods to reduce syntactic ambiguity caused by Prepositional Phrase (PP) attachment and coordination using semantic information. Our evaluation showed that clinical treebanks greatly improved the performance of existing parsers. The Berkeley parser achieved the best F-1 score of 86.39% on the MiPACQ treebank. For PP attachment, our proposed methods improved the accuracies of PP attachment by 2.35% on the MiPACQ corpus and 1.77% on the I2b2 corpus. For coordination, our method achieved a precision of 94.9% and a precision of 90.3% for the MiPACQ and i2b2 corpus, respectively. To further demonstrate the effectiveness of the improved parsing approaches, we applied outputs of our parsers to two external NLP tasks: semantic role labeling and temporal relation extraction. The experimental results showed that performance of both tasks’ was improved by using the parse tree information from our optimized parsers, with an improvement of 3.26% in F-measure for semantic role labelling and an improvement of 1.5% in F-measure for temporal relation extraction.
This study investigates the usefulness of the Treebank of Learner English (TLE) when applied to the task of Native Language Identification (NLI). The TLE is effectively a parallel corpus of Standard/Learner English, as there are two versions; one based on original learner essays, and the other an error-corrected version. We use the corpus to explore how useful a parser trained on ungrammatical relations is compared to a parser trained on grammatical relations, when used as features for a native language classification task. While parsing results are much better when trained on grammatical relations, native language classification is slightly better using a parser trained on the original treebank containing ungrammatical relations.
This paper gives an example of how enriched diachronic treebank data can shed new light on an old and conflicted topic, even when that topic is morphological and semantic in nature rather than syntactic. The topic is the rise of the Russian po delimitatives, a change seen as crucial in most accounts of the history of Russian aspect, since it represents a major step in generalising the derivational aspect system. Earlier accounts concur that the po delimitatives spread fairly recently, too recently for the development to be connected to the loss of the aorist tense, which also had delimitative readings with atelic verbs. Using treebank data from the Tromsø Old Russian and OCS Treebank, enriched with tags for derivational morphology and semantics, I show that the po delimitatives were not marginal even in the earliest Slavic sources, neither in terms of frequency nor semantics, and that they first complemented and then competed with the delimitative aorists. It can thus be claimed that the exotic po delimitatives grew organically out of the old Indo-European inflectional aspect system. The data set includes the data files and R scripts necessary to replicate the study.
FicTree is a dependency treebank of Czech fiction manually annotated in the format of the analytical layer of the Prague Dependency Trebank. The treebank consists of 12,760 sentences (166,432 tokens). The texts come from eight literary works published in the Czech Republic between 1991 and 2007. The syntactic annotation of the treebank was first performed by two distinct parsers (MSTParser and MaltParser) trained on the PDT training data, then manually corrected. Any differences between the two versions were resolved manually (by another annotator). The corpus is provided in a vertical format, where sentence boundaries are marked with a blank line. Every word form is written on a separate line, followed by five tab-separated attributes: lemma, tag, ID (word index in the sentence), head and deprel (analytical function, afun in the PDT formalism). The texts are shuffled in random chunks of maximum 100 words (respecting sentence boundaries). Each chunk is provided as a separate file, with the suggested division into train, dev and test sets written as file prefix.
This paper provides a new method to correct annotation errors in a treebank. The previous error correction method constructs a pseudo parallel corpus where incorrect partial parse trees are paired with correct ones, and extracts error correction rules from the parallel corpus. By applying these rules to a treebank, the method corrects errors. However, this method does not achieve wide coverage of error correction. To achieve wide coverage, our method adopts a different approach. In our method, we consider that if an infrequent pattern can be transformed to a frequent one, then it is an annotation error pattern. Based on a tree mining technique, our method seeks such infrequent tree patterns, and constructs error correction rules each of which consists of an infrequent pattern and a corresponding frequent pattern. We conducted an experiment using the Penn Treebank. We obtained 1,987 rules which are not constructed by the previous method, and the rules achieved good precision.
This paper describes the submission from the University of Helsinki to the\nshared task on cross-lingual dependency parsing at VarDial 2017. We present\nwork on annotation projection and treebank translation that gave good results\nfor all three target languages in the test set. In particular, Slovak seems to\nwork well with information coming from the Czech treebank, which is in line\nwith related work. The attachment scores for cross-lingual models even surpass\nthe fully supervised models trained on the target language treebank. Croatian\nis the most difficult language in the test set and the improvements over the\nbaseline are rather modest. Norwegian works best with information coming from\nSwedish whereas Danish contributes surprisingly little.\n
This article reflects on the notions of educated linguistic norm, standard linguistic norm and formal written mode in order to establish a productive dialogue with Portuguese teachers, especially those teaching and assessing their students’ written productions. We questioned some common procedures involved in the grading of written productions, such as focusing on grammar aspects of the texts based on obsolete grammar rules. In order to do so, we discussed the often overlooked and controversial senses associated with terms such as “educated” and “standard”, used interchangeably, for instance, by the graders of Brazil’s National High School Exam between 1998 and 2014 when assessing the written part of the exam. Later on, we performed a qualitative analysis of a school essay that presents a few particular morphosyntactic aspects that are in fact used by “educated” Brazilians in their daily and monitored writing practices. We sought to demonstrate that grammar tradition itself already often validates these aspects in their prescription. This way, possible interdictions in formal written contexts, such as in the National High School Exam, arise from an ideology that preaches unattainable standard linguistic rules, resulting in an obsolete, “pure” and overcorrect linguistic standard.Keywords: standard linguistic norm, grading, written production.
This poster will describe a collaboration between nursing facility staff and university researchers to establish individualized music interventions to improve targeted behavioral outcomes for nursing home residents with dementia. We used single-case design to assess intervention effectiveness for each resident. We will present data from 4–5 patients to illustrate how this type of design can be used to inform intervention targeting. We used independent sample t-tests to compare mean percentages of observed affect and behaviors from the Philadelphia Geriatric Center Affect Rating Scale and a modified Passivity in Dementia Scale before and during the music intervention and with a control period of attention only. Preliminary analyses show that for one resident, the music intervention did not produce significant differences in affect or behavior, but for a second resident positive affect (t(21) = -3.784, p =.001) was significantly higher during the intervention, and negative affect (t(19.674) = 2.595, p =.017) and negative behaviors (t(15.453) = 3.242, p =.005) were significantly lower during the intervention. We illustrate how reports from these simple analyses can be used to inform nursing home staff about how to use individualized music with residents who will most benefit from the intervention.
This study explores the question of whether native and non-native listeners, i.e. natives familiar with the language they are judging and non-natives who are not, manage to distinguish a foreign accent from a native accent in the speech of native speakers (NSs) and nonnative speakers (NNSs). Participants included 21 speakers (11 NSs and 10 NNSs who were native Turkish speakers) as well as two listener groups that consisted of 61 Finnish listeners (FLs), and 10 Turkish listeners (TLs) without Finnish experience. This study compares accent ratings by these two listener groups that evaluated the 21 spontaneous speech samples for foreign accent using a 9-point scale. The results showed a very significant difference between the listener groups for the NSs but no significant difference for the NNSs. The difference between the FL and the TL groups was because the FLs managed to distinguish the NSs from the NNSs, but otherwise these two listener groups exercised statistically similar ratings. Therefore, these results demonstrate that the listeners' familiarity with Finnish, the target language, hence listeners' native speaker status strongly affect ratings of foreign accents, since native listeners could distinguish the NSs, whereas non-native listeners could not. The results suggest that listeners' familiarity with the target language plays a much moreprofound role in accent detection than their familiarity with the accent language. Moreover, the results show that contrary to previous research, in the absence of listeners' familiarity with the target language, it is much more challenging to detect a foreign accent. The results also showed that speech rate correlated with the judgments provided by the TLs but not with the judgments provided by the FLs. This result raises the possibility that there are salient universal features of non-native speech such as speech rate that even non-native listeners unfamiliar with the language they are judging utilize while judging a foreign accent.
Study objective: The competitive sports environment can enhance social and cultural pressure towards having ideal body weight in weight-sensitive sports. The close relationship between body image and performance makes the elite athletes vulnerable to eating disorders. Thus, the purpose of this research was to study eating disorders and body image among weight-class elite athletes. Methods: A cross-sectional study was carried out with elite martial arts athletes (Karate, Taekwondo, and Judo) who were considered to be of higher risk for eating disorders. 63 elite martial arts male athletes (18.59 ± 5.29 yrs), and 63 non-athlete persons (17.3 ± 3.4 yrs) were recruited. Body Mass Index (BMI), Waist Hip Ratio (WHR), and Percent Body Fat (PBF) were measured using caliper and meter. Eating Disorder Diagnosis Scale (EDDS) and Body Image Rating Scale (BIRS) were used to study eating disorders and body image among elite martial arts athletes. Results: no sign of clinical EDDS were found among the investigated athletes, and non-athletes. There were significant differences in total score of EDDS (p=0.001), eating disorder and weight concern subscales (respectively p=0.012, p=0.001) in athletes and non-athletes. Furthermore, compared with the non-athlete group, elite athlete group with middle, good, and great body images scored higher on total score and all subscales of EDDS (p ≤ 0.05). Conclusion: The results from our study show the presence of worriment about eating disorder especially body weight and eating concern in elite athletes and the early detection of it may prevent progression to severe eating disorders.
Some attempts have been made in the academic community to carry out an automatic morphological analysis of the Qur'anic text. Among the well-known endeavors in this regard is the morphological annotation of the Quranic Arabic Corpus (QAC) which was carried out in Leeds University, UK. In addition, researchers in the University of Haifa had previously implemented a computational system for the morphological analysis of the Qur'an. More recently, a new Quranic corpus has been built in Mohammed I University in Morocco. To the best of our knowledge, these are the only three studies to produce a morphologically analyzed part-of-speech tagged Qur'an encoded as a structured linguistic database. This paper surveys the morphological analysis in the above-mentioned annotation projects and compares between them to test the quality of their analysis using five criteria related to display of the text in the corpus, word segmentation, morphological disambiguation, part of speech (POS) tag set and manual verification. The paper concludes that the QAC of Leeds and the Quranic corpus of Morocco surpass the Quranic corpus of Haifa with regard to most of these criteria. Furthermore, some additional POS tags for derivative nouns are suggested in a step to reach a more fine-grained tag set that could be proposed for POS tagging of Qur'anic Arabic.
<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
Precisely understanding the value and perception of consumers has long been recognized as essential elements of every market-oriented company's core business strategy. For this reason, customers' affection, as the basis for the formation of human values and judgment, should be considered carefully to strengthen the product quality and competitiveness. However, conventional product design places more attention to functional attributes and requires survey process to collect customers' evaluations, neglecting the in-depth study of the underlying associations between design properties and consumers' emotions based on the abundant online consumer response resources. To improve the deficiency, this study was proposed to develop a product affective properties identification approach. Particularly, data mining techniques (e.g. web mining, text mining) are applied to capture online product review resources. Considering the characteristics of user/consumer responses and evaluations, ontology is utilized to assist in the semantic analysis. With the help of product knowledge hierarchy and electronic lexical database, product properties, which can evoke consumers' affect, can be identified. Furthermore, the identified product affective properties are prioritized to provide designers with important reference for future improvement on the product. To illustrate the proposed approach, a pilot study based on iPhone 7 was conducted, in which the influential affective properties have been identified, and a ranking of them has been mapped out.
Laboratory work in biology teaching is an effective method to achieve learning objectives. In this practical activity presents about the role of plants, especially papaya fruit as an additional ingredient in the making of papaya bingka. This research aimed to know the students' response to laboratory work of making papaya bingka on sub chapter the role of plants in economics. Measurement of student's response to laboratory work of making papaya bingka using descriptive research. Data collection methods used responses questionnaire along with affective and psychomotor assessments with observation sheet, while cognitive assessments used student worksheet. Responses questionnaire used consisted of positive and negative statements of 12 statemets. The results of student's response showed that the average percentage of students' responses to laboratory work of making papaya bingka on submateri the role of plants in economic was 87% and categorized very high. The results was supported by affective rating of 87.16, psychomotor rating of 87.39 and cognitive rating of 88.70 values.
This paper suggests the hypothesis that emotion concepts which show similarity in their metaphorical mappings also reveal etymological interrelations across languages. In order to test this hypothesis, the Persian Linguistic Database was searched with keywords of two lexical fields of anger and fear. Thus 899 metaphorical expressions for anger and 600 ones for fear were found. By comparing the names of the mappings of anger and fear metaphors, we found out that these emotion concepts share 17 similar names of the mappings. The etymological data gathered from more than 25 languages for these fields also confirmed etymological and lexical interrelations. Hence, it can be concluded that there are correlations between conceptual metaphors and etymological and lexical patterns of fear and anger. The results of this research may not be limited to these two emotion concepts; it is suggested that such correlations between metaphorical mappings and etymological and lexical patterns can be found between other emotion concepts.
Affect prediction is a classical problem and has recently garnered special interest in multimedia applications. Affect prediction in movies is one such domain, potentially aiding the design as well as the impact analysis of movies. Given the large diversity in movies (such as different genres and languages), obtaining a comprehensive movie dataset for modeling affect is challenging while models trained on smaller datasets may not generalize. In this paper, we address the problem of continuous affect ratings with the availability of limited in-domain data resources. We initially setup several baseline models trained on in-domain data, followed by a proposal of a Knowledge Transfer (KT) + Gradient Boosting (GB) approach. KT learns models on a larger (mismatched) data which are then adapted to make predictions on the data of interest. GB further updates these predictions based on models learnt from the in-domain data. We observe that the KT + GB models provide Concordance Correlation Coefficient values of 0.13 and 0.27 for valence and affect prediction on the continuous LIRIS ACCEDE dataset against best baseline prediction values of 0.12 and 0.11. Not only the KT + GB models improve the overall performance metrics, we also observe a more consistent model performance across movies of various genres.
INTRODUCTION: Hoarding disorder can cause problems with work performance, personal hygiene, health and well-being. The disorder is a growing social problem in Japan. Having difficulty discarding rubbish, decluttering and organising can signal a future hoarding disorder, and early intervention is important. We developed an educational workshop on decluttering and organising for teens and adults with difficulty organising. The objective of this study is to evaluate the effectiveness of a workshop for reducing clutter and improving quality of life among younger people with difficulty decluttering and organising. METHODS AND ANALYSIS: An open-label, parallel-group, randomised controlled trial will be conducted among volunteers aged 12-55 years with mild difficulty decluttering and organising. Those in the intervention group will attend the workshop and receive a visit from a professional cleaning company to declutter their living space. The control group will have only the latter. The primary outcome will be the score on the Japanese version of the Saving Inventory-Revised. Secondary outcomes will be scores on the Clutter Image Rating Scale, the Japanese version of the Rosenberg Self-Esteem Scale and the Roles of Private Space Scale. The results will be examined for differences between the two groups in changes from baseline to 7 months. We will examine crude effects and adjust for gender and age using a general linear model for continuous variables and a logistic regression model for dichotomous variables. Sample size was calculated assuming a significance level of 5% (two tailed), a power of 80% and an effect size of 0.75. In total, 60 subjects (30 in each group) will be required. ETHICS AND DISSEMINATION: The study protocol has been approved by the Medical Ethical Committee of Teikyo University (No. 15-065). The findings will be disseminated widely through peer-reviewed publication and conference presentations. TRIAL REGISTRATION NUMBER: UMIN000020568. Issue date: 16 January 2016.
The practice of traditional medicine or herbal cures among the Bakossi people of Cameroon, as in any other African society, has been viewed with mixed feelings and sometimes with outright disdain; yet, it has its own type of achievements, particularly in the area of therapeutic cures or exorcism. That notwithstanding, in contemporary Cameroon society one notices how the advent of modern technological developments and Western education has greatly impacted the practice of traditional medicine and the way tangomas (medicine men) practise their art. This study investigates people’s possession of a sacred but rare ability to use language in a way peculiar to orthodox linguistic norms. It demonstrates the fact that oral literature has significant roles to play in African herbal medical practice. Furthermore, the paper examines disease taxonomy and how it affects language use among Bakossi tangomas. It is based on the premise that tangomas place high priority on the magical power of the word, as evidenced in their incantations, invocations and chants. The paper concludes that tangomas possess the best stock of poetical expressions and dramatic dialogue, which they employ to establish a rapport between people and the supernatural forces in the process of healing.
Most NLP resources that offer annotations at the word segment level provide morphological annotation that includes features indicating tense, aspect, modality, gender, case, and other inflectional information. Such information is rarely aligned to the relevant parts of the words-i.e. the allomorphs, as such annotation would be very costly. These unaligned weak labelings are commonly provided by annotated NLP corpora such as treebanks in various languages. Although they lack alignment information, the presence/absence of labels at the word level is also consistent with the amount of supervision assumed to be provided to L1 and L2 learners. In this paper, we explore several methods to learn this latent alignment between parts of word forms and the grammatical information provided. All the methods under investigation favor hypotheses regarding allomorphs of morphemes that re-use a small inventory, i.e. implicitly minimize the number of allomorphs that a morpheme can be realized as. We show that the provided information offers a significant advantage for both word segmentation and the learning of allomorphy.
While there is considerable evidence for neurobiological and neuropsychological impairments in clinical or incarcerated psychopaths, there is less research examining these domains among nonincarcerated adults with high levels of psychopathic traits (i.e., subclinical psychopaths).The present study examined differences in neuroendocrine and autonomic functioning, affective processing, and self-reported and behavioral aggression in young adult males with high levels of subclinical psychopathy, compared to young adult males with low levels of psychopathic characteristics.Participants completed selfreport questionnaires assessing psychopathic traits, aggression, drug and alcohol use, and other demographic information.They provided salivary cortisol samples before and 25 minutes after participating in a social stressor.Participants completed an affective processing task, during which their skin conductance reactivity was measured.They also engaged in a behavioral aggression task.Contrary to hypotheses, no differences between psychopathic groups were found in basal cortisol levels or cortisol reactivity following a social stressor, skin conductance reactivity while viewing negative pictures, valence ratings while viewing negative pictures, or aggression during a laboratory aggression task.Consistent with hypotheses, males with higher levels of psychopathic traits reported engaging in more physical and verbal aggression compared to the group with lower levels
The concepts of unitary evolution matrices and associative memory have boosted the field of Recurrent Neural Networks (RNN) to state-of-the-art performance in a variety of sequential tasks. However, RNN still have a limited capacity to manipulate long-term memory. To bypass this weakness the most successful applications of RNN use external techniques such as attention mechanisms. In this paper we propose a novel RNN model that unifies the state-of-the-art approaches: Rotational Unit of Memory (RUM). The core of RUM is its rotational operation, which is, naturally, a unitary matrix, providing architectures with the power to learn long-term dependencies by overcoming the vanishing and exploding gradients problem. Moreover, the rotational unit also serves as associative memory. We evaluate our model on synthetic memorization, question answering and language modeling tasks. RUM learns the Copying Memory task completely and improves the state-of-the-art result in the Recall task. RUM's performance in the bAbI Question Answering task is comparable to that of models with attention mechanism. We also improve the state-of-the-art result to 1.189 bits-per-character (BPC) loss in the Character Level Penn Treebank (PTB) task, which is to signify the applications of RUM to real-world sequential data. The universality of our construction, at the core of RNN, establishes RUM as a promising approach to language modeling, speech recognition and machine translation.
We explore the properties of byte-level recurrent language models. When given sufficient amounts of capacity, training data, and compute time, the representations learned by these models include disentangled features corresponding to high-level concepts. Specifically, we find a single unit which performs sentiment analysis. These representations, learned in an unsupervised manner, achieve state of the art on the binary subset of the Stanford Sentiment Treebank. They are also very data efficient. When using only a handful of labeled examples, our approach matches the performance of strong baselines trained on full datasets. We also demonstrate the sentiment unit has a direct influence on the generative process of the model. Simply fixing its value to be positive or negative generates samples with the corresponding positive or negative sentiment.
In this work, we present a minimal neural model for constituency parsing based on independent scoring of labels and spans. We show that this model is not only compatible with classical dynamic programming techniques, but also admits a novel greedy top-down inference algorithm based on recursive partitioning of the input. We demonstrate empirically that both prediction schemes are competitive with recent work, and when combined with basic extensions to the scoring model are capable of achieving state-of-the-art single-model performance on the Penn Treebank (91.79 F1) and strong performance on the French Treebank (82.23 F1).
We first present a minimal feature set for transition-based dependency parsing, continuing a recent trend started by Kiperwasser and Goldberg (2016a) and Cross and Huang (2016a) of using bi-directional LSTM features. We plug our minimal feature set into the dynamic-programming framework of Huang and Sagae (2010) and With our minimal features, we also present Opn 3 q global training methods. Finally, using ensembles including our new parsers, we achieve the best unlabeled attachment score reported (to our knowledge) on the Chinese Treebank and the "second-best-in-class" result on the English Penn Treebank.
We propose a neural network model for coordination boundary detection. Our method relies on the two common properties - similarity and replaceability in conjuncts - in order to detect both similar pairs of conjuncts and dissimilar pairs of conjuncts. The model improves identification of clause-level coordination using bidirectional RNNs incorporating two properties as features. We show that our model outperforms the existing state-of-the-art methods on the coordination annotated Penn Treebank and Genia corpus without any syntactic information from parsers.
In this paper, we attempt a comparison between new school transition-based parsers that use neural networks and their classical old school coun-terpart. We carry out experiments on treebanks fr...
Trance parsing model + Embedding vector. See https://github.com/tarowatanabe/trance for the parser and its usage. We also provide parsing and learning scripts for the Trance parser that we used for the paper; 1/ parsing model: ptb_train.txt.model-d100.tar.gz 2/ embedding vector: embedding-d100.vec.gz 3/ trance parser parsing script: trance-parsing.sh 4/ trance parser (batch) learning script: trance-training-batch.sh 5/ test.txt (gold file) and test.txt.leaf is for the parser input.
This study investigates the performance of 22 monolingual and 54 bilingual children with and without specific language impairment (SLI), in a nonword repetition (NWRT) and a sentence repetition task (SRT). Both tasks were constructed according to the principles for LITMUS tools (Language Impairment Testing in Multilingual Settings) developed within COST Action IS0804, and incorporated phonological or syntactic structures that are linguistically complex and have been shown to be difficult for children with SLI across languages. For phonology these are in particular (non)words containing consonant clusters. In morphosyntax complexity has been attributed to factors such as embedding and/or syntactic movement. Tasks focusing on such structures are expected to identify SLI in bilinguals across language combinations. This is notoriously difficult because structures that are problematic for typically developing bilinguals (BiTDs) and monolingual children with SLI (MoSLI) often overlap. We show that the NWRT and the SRT are reliable tools for identification of SLI in bilingual contexts. However, interpretation of the performance of bilingual children depends on background information as provided by parental questionnaires. In order to evaluate the accuracy of our tasks we recruited children in ordinary kindergartens or schools and in Speech Language Therapy centers and verified their status with a battery of standardized language tests, assessing bilingual children in both their languages. We consider a bilingual child language impaired if she shows impairments in two language domains in both her languages. For assessment we used tests normed for monolinguals (with one exception) and adjusted the norms for bilingualism and for language dominance. This procedure established the following groups: 10 typical monolinguals (MoTD), 12 MoSLI, 46 BiTD and 8 bilingual children with SLI (BiSLI). Our results show that both tasks target relevant structures: monolingual children are classified with 100% accuracy. Crucially, both our tasks distinguish BiTDs from MoSLIs and BiTDs from BiSLIs. The NWRT shows high accuracy and only minimal influence of language dominance. The SRT can be scored as “identical repetition” or as “target structure”, the latter aiming for scoring the mastery of a syntactic structure, ignoring lexical and specific case or gender errors. Focusing on the latter measure, we
The Ukrainian terminological system of architecture is a definitely organised, historically caused, communicative system of elements which functions, constantly changing internal and external communication at different levels of consistency. The terminological fund of representation of architecture knowledge demands settlement, therefore its research gets a special relevance. The national terminological system of architecture changes and updates constantly, because there are new concepts which are necessary to name, and there are new terms accordingly. Despite it, there is no clearness in formulation of definitions even for key concepts in this area of human activity, the choice of use and systematization of certain terminological units is not always successful, and synonymy of terms is actively used for base standard documents, the scientific special literature, lexicographic editions on architecture. Architecture terminology is a considerable layer of the Ukrainian terminological system with inherent common features for a system (presence of system elements, system signs, the criteria which are inherent for a system) and original Characteristics (specificity of concepts and terms’ relations, the special thematic and structural organisation which is the basis for allocation of architecture terminological system and allocates features of the language system organisation of(word-formation, grammatical, lexical). Elements of architectural terminological system form systems of objective realities, systems of terms, systems of corresponding concepts, systems of definitions of these concepts. When these systems function, they limit architectural terminology. It is very important to understand architectural terminology not simply as the mechanical sum of corresponding terminological units, terms and terminological word combinations, and accurate terminological system which has certain logic communications between its elements, structural components. There is a direct or indirect communication between terms which limits terminological system of architecture. Constant development and increase in quantity of terminological units, quantitative and qualitative changes of terms which unite in terminological system, define limitlessness of modern Ukrainian architecture terminological system. Exploring the practical problems of education terms and problems of architectural terms’ standardization the professional norm is fixed in the special literature (educational, methodical, statutory acts, state standards, scientific articles). Not always dictionaries in standard terminology consider professional terminological norm. The modern Ukrainian terminological fund of architecture makes specific, quantitatively big and qualitatively various layer of terminological lexicon in the typological characteristic of system. Architecture terminology is a component of Ukrainian terminological system with characteristic common features for system both the especial thematic and structural organisation which has differences in the language organisation of system.
The role of social media in giving voice to public opinion is impossible to ignore. Increasingly, platforms such as Facebook and Twitter are used for the mobilisation of action against public figures, corporations and organisations which have attracted negative public attention. For the social actors suffering this treatment, such actions may have devastating consequences.While previous studies have demonstrated that this mobilisation is in part the result of the real-time nature of social media, which allows for ‘rapid mass self-communication’ (Van der Meer & Verhoeven 2013) and the instant spreading of coherent frames across diverse groups of publics, the underlying conceptual dimension of these frames has only been studied to a limited degree, and primarily within crisis communication research (Ngai et al. 2015; Van der Meer et al. 2014). However, studying the conceptual grounding may offer additional and valuable explanations for the salience of particular frames and their ability to inspire collective action across different groups of publics. A previous, small-scale study indicates, for instance, that when commonly held notions of right and wrong are challenged, this leads to the establishment of strong and coherent frames that evoke socially and culturally embedded norms, and which not only have the purpose of condemning the culprit and his actions but also will unite publics in their call for corrective action (Author 2015). This paper reports on an explorative study that investigates the conceptual grounding of frames in instances of organisational and personal action that is deemed reproachful on social media. By examining a corpus of entries posted on Facebook in connection with two major organisational crises, the study confirms previous findings and demonstrates that the strength of frames may result from the evocation and foregrounding of basic social norms and values shared across public groups, which are otherwise considered to have different outlooks and perceptions.The theoretical foundation of the analysis is framing (Fillmore 1982; Hallahan 1999) combined with social media research (Liu 2010; Liu et al. 2011; Van der Meer & Verhoeven 2013), which provides the analyst with tools for investigating the conceptual and linguistic levels of communication on social media. Being concerned with the cognitive information processing of the receivers of text, framing can be instantiated through a number of lexical items, including metaphor (e.g. Lakoff and Johnson [1980]2003; Kövecses 2015). Due to its grounding in a bodily, situational, and discourse context as well as its richness in expression, metaphor is particularly relevant to this study and will receive special attention in the investigation of frames.
The paper investigates the third person singular pronoun anaphor e in Ga, a Kwa Language. The pronominal e which is prefixed may be glossed as ‘he’, ‘she’ or ‘it’. This pronominal which is prefixed refers to a definite third person discourse entity. It has no gender distinctions and it has also no animate distinctions. In Ga the pronominal e is attached to the verbs when in subjective position. When the pronominal is prefixed to the noun it marks possessive. However there are certain contexts in which the interpretation of the pronominal can be ‘he’ only or ‘she’ only and other instances cannot be either of them The aim is to find how the singular pronoun anaphor could be disambiguated in discourse especially where giving the referent to the anaphor is not simple. The paper applies the Centering Theory to analyse utterances in which this pronoun occurs and also the Givenness Hierarchy was considered in some cases where the referent resolution process involves a choice between two candidate referents. Data used was elicited by the researchers after having several chats with other natives of Ga. It came to light that in addition to the theories applied, the sociocultural norms and the lexical verb give clues in resolution of the third person singular pronoun. Keywords: pronominal, anaphor, centering theory, disambiguate, referent
We propose a question answering (QA) approach for standardized science exams that both identifies correct answers and produces compelling human-readable justifications for why those answers are correct. Our method first identifies the actual information needed in a question using psycholinguistic concreteness norms, then uses this information need to construct answer justifications by aggregating multiple sentences from different knowledge bases using syntactic and lexical information. We then jointly rank answers and their justifications using a reranking perceptron that treats justification quality as a latent variable. We evaluate our method on 1,000 multiple-choice questions from elementary school science exams, and empirically demonstrate that it performs better than several strong baselines, including neural network approaches. Our best configuration answers 44% of the questions correctly, where the top justifications for 57% of these correct answers contain a compelling human-readable justification that explains the inference required to arrive at the correct answer. We include a detailed characterization of the justification quality for both our method and a strong baseline, and show that information aggregation is key to addressing the information need in complex questions.
Since the age of paper versions, dictionaries are often published with anomalies in their content resulting from lexicographer’s mistakes or from the lack of efficiency of automatic enrichment systems. Many of these anomalies are expensive to manually detect and difficult to automatically control, notably with lightly structured models of dictionaries. In this article, we take advantage of the fine structure proposed by the Lexical Markup Framework (LMF) norm to investigate the detection of anomalies in the content of LMF normalized dictionaries. First, we give a theoretical study on the plausible anomalies, such as inconsistency, incoherence, redundancy, and incompleteness. Second, we detail the approach that we propose for the automatic detection of such anomalies. Finally, we report on an experiment carried out on an available normalized dictionary of the Arabic language. The experiment has shown that the proposed approach gives reasonable results in terms of precision and recall.
Text-setting, the arrangement of language to music, is a common source of evidence in the debate over the relevance of the syllable in Japanese prosody (e.g., Labrune 2012). Although Japanese text-setting is typically treated as mora-based, the present corpus analysis reveals that syllable-based text-setting is pervasive in Japanese. Two studies presented here compare native Japanese songs with those translated into Japanese. The results demonstrate use of syllabic settings throughout the corpora and across the lexical strata of Japanese. Syllabic settings are shown to arise with greater likelihood in response to pressures imposed by restrictive translation contexts, information density mismatch, and knowledge of correspondence to English loans. We argue that, given the viability of syllabic text-setting in Japanese, moraic text-setting is a stylistic norm of Japanese music that is shifting over time, rather than evidence of a lack of syllable structure in the language’s prosodic system.
Researchers have recently introduced various LexTALE-type word recognition tests in order to assess vocabulary size in a second language (L2) mastered by participants. These tests correlate well with other measures of language proficiency in unbalanced bilinguals whose second language ability is well below the level of their native language. In the present study, we investigated whether LexTALE-type tests also discriminate at the high end of the proficiency range. In several regions of Spain, people speak both the regional language (e.g., Catalan or Basque) and Spanish to very high degrees. Still, because of their living circumstances, some consider themselves as either Spanish-dominant or regional-language dominant. We showed that these two groups perform differently on the recently published Spanish Lextale-Esp: The Spanish-dominant group had significantly higher scores than the Catalan-dominant group. We also showed that the noncognate words of the test have the highest discrimination power. This indicates that the existing Lextale-Esp can be used to estimate proficiency differences in highly proficient bilinguals with Spanish as an L2, and that a more sensitive test could be built by replacing the cognates.
The present article considers poem as a narrative of language and, based on the ideas of Russian formalists (literary highlighting and linguistic norm-evasion), analyzes the lexical and semantic level of female poets’ poems in pre-contemporary periods (Parvin Etesami), contemporary (Forough Farrokhzad) and the revolutionary period (Fatemeh Rake'i), then examines the gender variable in the vocabulary choice and the meaning of their poems. The results are as follows: the vocabulary used in Parvin’s poems refers directly to her feminine viewpoint generally, although her maternal poems both in terms of words and meaning imply her womanhood and her motherhood. Forough uses words which are sexual meaning laden, and man-woman dichotomy analysis is prevalent semantically in her poems. The femininity of Rakei’s poems is evident in feminine vocabularies and feminine names she has employed in her poems. Semantically, motherhood is the dominant aspect of her poems. She has developed this dimension and has universalized motherhood sense in her poetries. The research method of this paper is a qualitative of the content analysis kind. It has been shown that femininity has been influential not only in standard language, but also in literary language.
Sleep consolidates memory for emotional events. Whether it likewise preserves or, to the contrary, depotentiates emotional responses associated with such events is unclear, as there is evidence to support both cases. Conflicting findings may be partially due to differences in the way emotional reactivity is measured. Thus, the goal of the current project is to investigate the influence of sleep on emotional reactivity using multiple simultaneous measures. Young adults (18–30 years) viewed 45 negative and 45 neutral pictures before a period containing either a nap measured with polysomnography (n=33) or wake (n=25). Following the nap or wake period, they viewed the same pictures intermixed with novel ones and indicated whether they remembered each picture. Emotional response to each picture was measured at both time points with valence and arousal ratings, skin conductance response, heart rate deceleration, and corrugator supercilii electromyography (corrugator EMG). Across the sessions, both valence ratings (t=2.043, p=0.046) and skin conductance response (t=-4.052, p<0.001) were preserved for negative pictures in the nap group relative to the wake group. Arousal ratings, heart rate deceleration, and corrugator EMG response declined non-significantly more in the wake group than the nap group. Participants in the nap group remembered non-significantly more pictures than those in the wake group (M+/-SD: 85.25+/-10.14% vs. 80.83+/-10.56%). Polysomnography data will be analyzed to investigate the relationship between specific features of sleep and change in reactivity measures. These data suggest that sleep consolidates reactivity associated with emotional experiences whether it is measured with subjective rating scales or objective physiological responses. This work was funded by NIH R01 AG040133 (PI: Spencer).