Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
16504 papers
Stanford Dependencies (SD) represent nowadays a de facto standard as far as dependency annotation is concerned. The goal of this paper is to explore pros and cons of different strategies for generating SD annotated Italian texts to enrich the existing Italian Stanford Dependency Treebank (ISDT). This is done by comparing the performance of a statistical parser (DeSR) trained on a simpler resource (the augmented version of the Merged Italian Dependency Treebank or MIDT+) and whose output was automatically converted to SD, with the results of the parser directly trained on ISDT. Experiments carried out to test reliability and effectiveness of the two strategies show that the performance of a parser trained on the reduced dependencies repertoire, whose output can be easily converted to SD, is slightly higher than the performance of a parser directly trained on ISDT. A non-negligible advantage of the first strategy for generating SD annotated texts is that semi-automatic extensions of the training resource are more easily and consistently carried out with respect to a reduced dependency tag set. Preliminary experiments carried out for generating the collapsed and propagated SD representation are also reported.
Previous work by Lin et al. (2011) demonstrated the effectiveness of using discourse relations for evaluating text coherence. However, their work was based on discourse relations annotated in accordance with the Penn Discourse Treebank (PDTB) (Prasad et al., 2008), which encodes only very shallow discourse structures; therefore, they cannot capture long-distance discourse dependencies. In this paper, we study the impact of deep discourse structures for the task of co-herence evaluation, using two approaches: (1) We compare a model with features derived from discourse relations in the style of Rhetorical Structure Theory (RST) (Mann and Thompson, 1988), which annotate the full hierarchical discourse structure, against our re-implementation of Lin et al.’s model; (2) We compare a model encoded using only shallow RST-style discourse relations, against the one encoded using the complete set of RST-style discourse relations. With an evaluation on two tasks, we show that deep discourse structures are truly useful for better dif-ferentiation of text coherence, and in general, RST-style encoding is more powerful than PDTB-style encoding in these settings. 1
The neuropeptide oxytocin enhances in-group favoritism and ethnocentrism in males. However, whether such effects also occur in women and extend to national symbols and companies/consumer products is unclear. In a between-subject, double-blind placebo controlled experiment we have investigated the effect of intranasal oxytocin on likeability and arousal ratings given by 51 adult Chinese males and females for pictures depicting people or national symbols/consumer products from both strong and weak in-groups (China and Taiwan) and corresponding out-groups (Japan and South Korea). To assess duration of treatment effects subjects were also re-tested after 1 week. Results showed that although oxytocin selectively increased the bias for overall liking for Chinese social stimuli and the national flag, it had no effect on the similar bias toward other Chinese cultural symbols, companies, and consumer products. This enhanced bias was maintained 1 week after treatment. No overall oxytocin effects were found for Taiwanese, Japanese, or South Korean pictures. Our findings show for the first time that oxytocin increases liking for a nation's society and flag in both men and women, but not that for other cultural symbols or companies/consumer products.
Studies comparing memory and future event simulation find that future events are more positive, and more often depend on life script events (e.g., culturally normative landmark events) than past events. Previous research does not address the link between this positivity bias and the life stage of college-age participants or their reliance on these scripted events. To examine this positivity bias, narratives of past and anticipated future events were elicited from participants aged 18-74 years, and were examined for reliance on the life script and valence ratings. Results showed that, across age groups, future events were rated as more positive than past events, and that life script events were common in the distant future. Notably, whereas younger adult age groups wrote primarily about their own life script events, older participants more commonly wrote about attending the life script events of significant others, such as children and grandchildren. These findings suggest that simulated future events play a valuable role in self-enhancement across the lifespan. Furthermore, the life script can be viewed as a useful search mechanism when one is missing the episodic details that are more available in memories; however, it is not the source of positivity bias for future events.
We present a user-centered approach for defining the dependency syntactic specification for a treebank. We show that by collecting information on syntactic interpretations from the future users of the treebank, we can model so far dependency-syntactically undefined syntactic structures in a way that corresponds to the users’ intuition. By consulting the users at the grammar definition phase we aim at better usage of the treebank in the future. We focus on two complex syntactic phenomena: elliptical comparative clauses and participial NPs or NPs with a verb-derived noun as their head. We show how the phenomena can be interpreted in several ways and ask for the users’ intuitive way of modeling them. The results aid in constructing the syntactic specification for the treebank.
Social media applications such as Twitter provide a powerful medium through which users can communicate their observations with friends and with the world at large. We have witnessed live reporting of many events, from soccer games in Johannesburg to revolutions in Cairo and Tunis, and these reports have in many ways rivaled the content provided by the official media. Tapping into this valuable resource is a challenge, due to the heterogeneity and noise inherent in realtime text, diversity of languages, and fast-evolving linguistic norms. In this paper we seek to analyze a tweet stream to automatically discover points in time when an important event happens, and to classify such events based on the type of the sentiments they evoke, using only non-textual features of the tweeting pattern. This results not only in a robust way of analyzing tweet streams independent of the languages used; it also provides insights about how users behave on social media websites. For example, we observe that users often react to an exciting external event by decreasing the volume of communication with other users. We explain this effect through a model of how users switch between producing information or sentiments and sharing others’ news or sentiments. We develop and evaluate our models and algorithms using several Twitter data sets, focusing in particular on the tweets sent during the soccer World Cup of 2010. This data set has the feature that the underlying ground truth is welldefined and known whereby goals serve as events.
Abstract This study was conducted to understand the relationship between familiarity and cross‐cultural acceptance for an ethnic sweet treat ( Y ackwa; K orean traditional cookie) by K orean, J apanese and F rench consumers. Descriptive analysis and consumer testing were performed on six Y ackwa samples. Overall, the samples received favorable responses from the foreign consumers. K orean consumers liked samples with a soft and cohesive texture, whereas J apanese and F rench consumers liked flaky and crispy texture. French consumers rated stronger sweetness to be more appropriate for Y ackwa compared to K orean and J apanese consumers. Texture liking was strongly correlated with familiarity rating in all three countries, indicating that the consumers' previous experience with similar products might affect their preference for certain textural attributes. Familiarity was correlated with all hedonic ratings by K orean consumers, who are most familiar with Y ackwa, but with overall and texture liking by J apanese consumers and flavor and texture liking by French consumers. These results suggest that familiarity partly contributes to a foreign consumers' hedonic rating. Practical Applications Globalization and cultural diversity have increased interest in ethnic foods. This trend is motivating food industries to expand into the ethnic food market sector. In this study, the sensory attributes and the cross‐cultural acceptability of Y ackwa ( K orean traditional cookie) were evaluated and the potential role of familiarity in determining consumer acceptance was measured. The outcome of this study will help food exporters, R&D scientists and food marketers in ethnic food market to optimize an ethnic food for other cultural communities by educating them to consider familiarity as an important factor for product development and promotion.
BACKGROUND: Depression is frequently characterized by patterns of inflexible, maladaptive, and ruminative thinking styles, which are thought to result from a combination of decreased attentional control, decreased executive functioning, and increased negative affect. Cognitive Control Training (CCT) uses computer-based behavioral exercises with the aim of strengthening cognitive and emotional functions. A previous study found that severely depressed participants who received CCT exhibited reduced negative affect and rumination as well as improved concentration. AIMS: The present study aimed to extend this line of research by employing a more stringent control group and testing the efficacy of three sessions of CCT over a 2-week period in a community population with depressed mood. METHOD: Forty-eight participants with high Beck Depression Inventory (BDI-II) scores were randomized to CCT or a comparison condition (Peripheral Vision Training; PVT). RESULTS: Significant large effect sizes favoring CCT over PVT were found on the BDI-II (d = 0.73, p <.05) indicating CCT was effective in reducing negative mood. Additionally, correlations showed significant relationships between CCT performance (indicating ability to focus attention on CCT) and state affect ratings. CONCLUSIONS: Our results suggest that CCT is effective in altering depressed mood, although it may be specific to select mood dimensions.
Sentiment analysis has now become a popular research problem to tackle in NLP field. However, there are very few researches conducted on sentiment analysis for Chinese. Progress is held back due to lack of large and labelled corpus and powerful models. To remedy this deficiency, we build a Chinese Sentiment Treebank over social data. It concludes 13550 labeled sentences which are from movie reviews. Furthermore, we introduce a novel Recursive Neural Deep Model (RNDM) to predict sentiment label based on recursive deep learning. We consider the problem of classifying one sentence by overall sentiment, determining a review is positive or negative. On predicting sentiment label at sentence level, our model outperforms other commonly used baselines, such as Naïve Bayes, Maximum Entropy and SVM, by a large margin.
In adults, patterns of neural activation associated with perhaps the most basic language skill--overt object naming--are extensively modulated by the psycholinguistic and visual complexity of the stimuli. Do children's brains react similarly when confronted with increasing processing demands, or they solve this problem in a different way? Here we scanned 37 children aged 7-13 and 19 young adults who performed a well-normed picture-naming task with 3 levels of difficulty. While neural organization for naming was largely similar in childhood and adulthood, adults had greater activation in all naming conditions over inferior temporal gyri and superior temporal gyri/supramarginal gyri. Manipulating naming complexity affected adults and children quite differently: neural activation, especially over the dorsolateral prefrontal cortex, showed complexity-dependent increases in adults, but complexity-dependent decreases in children. These represent fundamentally different responses to the linguistic and conceptual challenges of a simple naming task that makes no demands on literacy or metalinguistics. We discuss how these neural differences might result from different cognitive strategies used by adults and children during lexical retrieval/production as well as developmental changes in brain structure and functional connectivity.
Although exposure therapy is an effective treatment for anxiety disorders, fear sometimes returns following successful therapy. The Rescorla–Wagner model predicts that presenting two fear-provoking stimuli simultaneously (compound extinction) will maximize learning during exposure and reduce the likelihood of relapse. Participants were presented with either single extinction trials only or single extinction trials followed by compound extinction trials. In addition, participants within each extinction group were randomized to caffeine or placebo ingestion prior to extinction to investigate the mechanism by which compound extinction may maximize learning (enhanced associative change or enhanced responding). Participants presented with compound trials demonstrated significantly less fear responding at spontaneous recovery compared with participants who received single extinction trials only. Ingestion of caffeine also provided some protection from spontaneous recovery (as measured by valence ratings). At the reinstatement test, only compound extinction trials predicted less fear responding; caffeine ingestion prior to extinction did not attenuate reinstatement effects.
Our first impression of others is highly influenced by their facial appearance. However, the perception and evaluation of faces is not only guided by internal features such as facial expressions, but also highly dependent on contextual information such as secondhand information (verbal descriptions) about the target person. To investigate the time course of contextual influences on cortical face processing, event-related brain potentials were investigated in response to neutral faces, which were preceded by brief verbal descriptions containing cues of affective valence (negative, neutral, positive) and self-reference (self-related vs. other-related). ERP analysis demonstrated that early and late stages of face processing are enhanced by negative and positive as well as self-relevant descriptions, although faces per se did not differ perceptually. Affective ratings of the faces confirmed these findings. Altogether, these results demonstrate for the first time both on an electrocortical and behavioral level how contextual information modifies early visual perception in a top-down manner.
Recent work on Chinese analysis has led to large-scale annotations of the internal structures of words, enabling characterlevel analysis of Chinese syntactic structures. In this paper, we investigate the problem of character-level Chinese dependency parsing, building dependency trees over characters. Character-level information can benefit downstream applications by offering flexible granularities for word segmentation while improving wordlevel dependency parsing accuracies. We present novel adaptations of two major shift-reduce dependency parsing algorithms to character-level parsing. Experimental results on the Chinese Treebank demonstrate improved performances over word-based parsing methods.
Prior research suggests that repeatedly approaching or avoiding a certain stimulus changes the liking of this stimulus. We investigated whether these effects of approach and avoidance training occur also when participants do not perform these actions but are merely instructed about the stimulus-action contingencies. Stimulus evaluations were registered using both implicit (Implicit Association Test and evaluative priming) and explicit measures (valence ratings). Instruction-based approach-avoidance effects were observed for relatively neutral fictitious social groups (i.e., Niffites and Luupites), but not for clearly valenced well-known social groups (i.e., Blacks and Whites). We conclude that instructions to approach or avoid stimuli can provide sufficient bases for establishing both implicit and explicit evaluations of novel stimuli and discuss several possible reasons for why similar instruction-based approach-avoidance effects were not found for valenced well-known stimuli.
In this paper, we present our work of humor recognition on Twitter, which will facilitate affect and sentimental analysis in the social network. The central question of what makes a tweet (Twitter post) humorous drives us to design humor-related features, which are derived from influential humor theories, linguistic norms, and affective dimensions. Using machine learning techniques, we are able to recognize humorous tweets with high accuracy and F-measure. More importantly, we single out features that contribute to distinguishing non-humorous tweets from humorous tweets, and humorous tweets from other short humorous texts (non-tweets). This proves that humorous tweets possess discernible characteristics that are neither found in plain tweets nor in humorous non-tweets. We believe our novel findings will inform and inspire the burgeoning field of computational humor research in the social media.
Neural substrates underlying the human-pet relationship are largely unknown. We examined fMRI brain activation patterns as mothers viewed images of their own child and dog and an unfamiliar child and dog. There was a common network of brain regions involved in emotion, reward, affiliation, visual processing and social cognition when mothers viewed images of both their child and dog. Viewing images of their child resulted in brain activity in the midbrain (ventral tegmental area/substantia nigra involved in reward/affiliation), while a more posterior cortical brain activation pattern involving fusiform gyrus (visual processing of faces and social cognition) characterized a mother's response to her dog. Mothers also rated images of their child and dog as eliciting similar levels of excitement (arousal) and pleasantness (valence), although the difference in the own vs. unfamiliar child comparison was larger than the own vs. unfamiliar dog comparison for arousal. Valence ratings of their dog were also positively correlated with ratings of the attachment to their dog. Although there are similarities in the perceived emotional experience and brain function associated with the mother-child and mother-dog bond, there are also key differences that may reflect variance in the evolutionary course and function of these relationships.
BACKGROUND: Music can elicit strong emotions and can be remembered in connection with these emotions even decades later. Yet, the brain correlates of episodic memory for highly emotional music compared with less emotional music have not been examined. We therefore used fMRI to investigate brain structures activated by emotional processing of short excerpts of film music successfully retrieved from episodic long-term memory. METHODS: Eighteen non-musicians volunteers were exposed to 60 structurally similar pieces of film music of 10 s length with high arousal ratings and either less positive or very positive valence ratings. Two similar sets of 30 pieces were created. Each of these was presented to half of the participants during the encoding session outside of the scanner, while all stimuli were used during the second recognition session inside the MRI-scanner. During fMRI each stimulation period (10 s) was followed by a 20 s resting period during which participants pressed either the "old" or the "new" button to indicate whether they had heard the piece before. RESULTS: Musical stimuli vs. silence activated the bilateral superior temporal gyrus, right insula, right middle frontal gyrus, bilateral medial frontal gyrus and the left anterior cerebellum. Old pieces led to activation in the left medial dorsal thalamus and left midbrain compared to new pieces. For recognized vs. not recognized old pieces a focused activation in the right inferior frontal gyrus and the left cerebellum was found. Positive pieces activated the left medial frontal gyrus, the left precuneus, the right superior frontal gyrus, the left posterior cingulate, the bilateral middle temporal gyrus, and the left thalamus compared to less positive pieces. CONCLUSION: Specific brain networks related to memory retrieval and emotional processing of symphonic film music were identified. The results imply that the valence of a music piece is important for memory performance and is recognized very fast.
In the context of Internet addiction, cybersex is considered to be an Internet application in which users are at risk for developing addictive usage behavior. Regarding males, experimental research has shown that indicators of sexual arousal and craving in response to Internet pornographic cues are related to severity of cybersex addiction in Internet pornography users (IPU). Since comparable investigations on females do not exist, the aim of this study is to investigate predictors of cybersex addiction in heterosexual women. We examined 51 female IPU and 51 female non-Internet pornography users (NIPU). Using questionnaires, we assessed the severity of cybersex addiction in general, as well as propensity for sexual excitation, general problematic sexual behavior, and severity of psychological symptoms. Additionally, an experimental paradigm, including a subjective arousal rating of 100 pornographic pictures, as well as indicators of craving, was conducted. Results indicated that IPU rated pornographic pictures as more arousing and reported greater craving due to pornographic picture presentation compared with NIPU. Moreover, craving, sexual arousal rating of pictures, sensitivity to sexual excitation, problematic sexual behavior, and severity of psychological symptoms predicted tendencies toward cybersex addiction in IPU. Being in a relationship, number of sexual contacts, satisfaction with sexual contacts, and use of interactive cybersex were not associated with cybersex addiction. These results are in line with those reported for heterosexual males in previous studies. Findings regarding the reinforcing nature of sexual arousal, the mechanisms of learning, and the role of cue reactivity and craving in the development of cybersex addiction in IPU need to be discussed.
We present a novel approach for induc-ing unsupervised dependency parsers for languages that have no labeled training data, but have translated text in a resource-rich language. We train probabilistic pars-ing models for resource-poor languages by transferring cross-lingual knowledge from resource-rich language with entropy reg-ularization. Our method can be used as a purely monolingual dependency parser, requiring no human translations for the test data, thus making it applicable to a wide range of resource-poor languages. We perform experiments on three Data sets — Version 1.0 and version 2.0 of Google Universal Dependency Treebanks and Treebanks from CoNLL shared-tasks, across ten languages. We obtain state-of-the art performance of all the three data sets when compared with previously studied unsupervised and projected pars-ing systems. 1
Why do some people like negative, or even disgusting and provocative artworks? Art expertise, believed to influence the interplay among cognitive and emotional processing underlying aesthetic experience, could be the answer. We studied how art expertise modulates the effect of positive-and negative-valenced artworks on aesthetic and emotional responses, measured with self-reports and facial electromyography (EMG). Unsurprisingly, emotionally-valenced art evoked coherent valence as well as corrugator supercilii and zygamoticus major activations. However, compared to non-experts, experts showed attenuated reactions, with less extreme valence ratings and corrugator supercilii activations and they liked negative art more. This pattern was also observed for a control set of International Affective Picture System (IAPS) pictures suggesting that art experts show general processing differences for visual stimuli. Thus, much in line with the Kantian notion that an aesthetic stance is emotionally distanced, art experts exhibited a distinct pattern of attenuated emotional responses.
We describe a new dependency parser for English tweets, TWEEBOPARSER. The parser builds on several contributions: new syntactic annotations for a corpus of tweets (TWEEBANK), with conventions informed by the domain; adaptations to a statistical parsing algorithm; and a new approach to exploiting out-of-domain Penn Treebank data. Our experiments show that the parser achieves over 80% unlabeled attachment accuracy on our new, high-quality test set and measure the benefit of our contributions. Our dataset and parser can be found at http://www.ark.cs.cmu.edu/TweetNLP.
OBJECTIVE: In alcohol-dependent patients, alcohol cues evoke increased activation in mesolimbic brain areas, such as the nucleus accumbens and the amygdala. Moreover, patients show an alcohol approach bias, a tendency to more quickly approach than avoid alcohol cues. Cognitive bias modification training, which aims to retrain approach biases, has been shown to reduce alcohol craving and relapse rates. The authors investigated effects of this training on cue reactivity in alcohol-dependent patients. METHOD: In a double-blind randomized design, 32 abstinent alcohol-dependent patients received either bias modification training or sham training. Both trainings consisted of six sessions of the joystick approach-avoidance task; the bias modification training entailed pushing away 90% of alcohol cues and 10% of soft drink cues, whereas this ratio was 50/50 in the sham training. Alcohol cue reactivity was measured with functional MRI before and after training. RESULTS: Before training, alcohol cue-evoked activation was observed in the amygdala bilaterally, as well as in the right nucleus accumbens, although here it fell short of significance. Activation in the amygdala correlated with craving and arousal ratings of alcohol stimuli; correlations in the nucleus accumbens again fell short of significance. After training, the bias modification group showed greater reductions in cue-evoked activation in the amygdala bilaterally and in behavioral arousal ratings of alcohol pictures, compared with the sham training group. Decreases in right amygdala activity correlated with decreases in craving in the bias modification but not the sham training group. CONCLUSIONS: These findings provide evidence that cognitive bias modification affects alcohol cue-induced mesolimbic brain activity. Reductions in neural reactivity may be a key underlying mechanism of the therapeutic effectiveness of this training.
According to Tsinghua Chinese Treebank annotation methods, the authors extracted relation words and marked their categories. Then syntax, lexical and position features of automatic syntax tree with and without functional marker were extracted to recognize and classify relation words. Experiment results show that relative recognition accuracy is 95.7%, and relation words classification F1 is 77.2%.
In this paper, we analyze the impact of various dependency representations for various constructions on the general parsing accuracy and on the parsing accuracy of these constructions. We focus on the analysis of coordination constructions, complex predicates, and punctuation mark attachment. We use Latvian Treebank as a dataset, thus, providing insight for an inflective language with a rather free word order. Experiments with MaltParser, a transition-based parser, show clear difference in learnability of various representations for the considered constructions. Future work would include carrying out comparable experiments with a graph-based dependency parser like MSTParser.
In Dutch V-final clauses the verbs tend to form a cluster which cannot be split up by nonverbal \nmaterial. However, Haeseryn et al. (1997) as well as other studies on the phenomenon list several \ncases in which the verb cluster may be interrupted by \ncluster creepers. \nThe most common examples are constructions with separable verb particles, but examples with nouns, adjectives, and adverbs are attested as well. \nSince the majority of the data in previous studies is collected by introspection and elicitation, \nit is interesting to compare those findings to corpus data. The corpus analysis is based on data \nfrom two Dutch treebanks (CGN and LASSY), which allow to take into account regional and/or \nstylistic variation. This is an important aspect for the analysis, since cluster creeping is reported \nto be a typical property of spoken and regional variants of Dutch. \nThe goal of this corpus-based investigation is on the one hand to provide insight in the frequency \nof the phenomenon, and on the other hand to classify the types of cluster creepers. Besides the \nlinguistic analysis, methodological issues regarding the extraction of the relevant data from the \ntreebanks will be addressed as well.
The paper tries to contribute to the general discussion on discourse connectives, concretely to the question whether it is meaningful to distinguish two separate groups of connectives -i.e."classical" connectives limited to few predefined classes like conjunctions or adverbs (e.g.but) vs. alternative lexicalizations of connectives (i.e.unrestricted expressions and phrases like the reason is, he added, the condition was etc.).In this respect, the paper focuses on one group of these broader connectives in Czech -the selected verbs of saying doplnit/doplňovat (to complement), upřesnit/upřesňovat (to specify), dodat/dodávat (to add), pokračovat (to continue) -and analyses their occurrence and function in texts from the Prague Discourse Treebank.The paper demonstrates that these verbs of saying have a special place within the other connectives, as they contain two items -e.g. he added means and he said so the verb to add contains an information about the relation to the previous context (and) plus the verb of saying (to say).This information led us to a more general observation, i.e. discourse connectives in broader sense do not necessarily connect two pieces of a text but some of them carry the second argument right in their semantics, which "classical" connectives can never do.
The paper introduces a possibility of new research offered by a multi-dimensional annotation of the Prague Dependency Treebank. It focuses on exploitation of the annotation of coreference for the annotation of discourse relations expressed by multiword expressions. It tries to find which as-pect interlinks these linguistic areas and how we can use this interplay in automatic searching for Czech expressions like despite this (navzdory tomu), because of this fact (díky této skutečnosti) functioning as multiword discourse markers. 1
We describe and evaluate the semi-automatic addition of a deep syntactic layer to the French Treebank (Abeillé and Barrier [1]), using an existing scheme (Candito et al. [6]). While some rare or highly ambiguous deep phenomena are handled manually, the remainings are derived using a graph-rewriting system (Ribeyre et al. [22]). Although not manually corrected, we think the resulting Deep Representations can pave the way for the emergence of deep syntactic parsers for French.
While annotated treebanks are an invaluable tool in linguistic research, the tree-based form in which corpus search tools often present search results is not necessarily well-suited to the user’s requirements. We argue that a concordance style export of search results, built around a user-identified "key node" in the query, represents a useful synoptic view of the data for the user needing to carry out further manual analysis of query results. We present a first implementation of these 'KNIC' concordances for an Old French corpus, using the TigerSearch treebank search engine integrated into the TXM corpus analysis platform.
This paper mainly introduced the research on constructing Mongolian Treebank based on phrase structure grammar. Having Considered related Mongolian Treebank work and Mongolian words characteristics, we developed a Mongolian syntactic tagset. The tagset includes two kinds of tags. One is syntactic constituent tag and the other is grammatical relation tag. On the basis of the tagset, we developed the Mongolian Treebank auxiliary processing system. Finally, we built a Treebank that contains 3645 sentences and did an experiment on this Treebank.
A relatively new theory of motivation posits that purposeful human behavior may be partly explained by multidimensional individual differences "traits of action" (motives). Its 15 motives can be characterized according to their purpose: individual integrity, competitiveness, and cooperativeness. Existing evidence supports the model on which the motives are based and the reliability and validity of strategies to assess them. This experiment tested whether the hypothetical results of consistent, motivated cooperative and competitive behavior could affect ratings of attractiveness. Male and female participants (N = 98; M age = 18.8, SD = 1.4) were shown 24 opposite-sex facial photos ranging in attractiveness. The photos were paired with one of three conditions representing theoretical outcomes that would result from low, control, and high levels of cooperative and competitive motives. As predicted, outcome descriptions representing high motive strength of six motives statistically significantly affected ratings of attractiveness. This result was independent of sex of participant and consistent with the theory.
Sentiment analysis of short texts such as single sentences and Twitter messages is challenging because of the limited contextual information that they normally contain. Effectively solving this task requires strategies that combine the small text content with prior knowledge and use more than just bag-of-words. In this work we propose a new deep convolutional neural network that exploits from characterto sentence-level information to perform sentiment analysis of short texts. We apply our approach for two corpora of two different domains: the Stanford Sentiment Treebank (SSTb), which contains sentences from movie reviews; and the Stanford Twitter Sentiment corpus (STS), which contains Twitter messages. For the SSTb corpus, our approach achieves state-of-the-art results for single sentence sentiment prediction in both binary positive/negative classification, with 85.7% accuracy, and fine-grained classification, with 48.3% accuracy. For the STS corpus, our approach achieves a sentiment prediction accuracy of 86.4%.
This article presents an ensemble parse approach to detecting and selecting high-quality linguistic analyses output by a hand-crafted HPSG grammar of Spanish implemented in the LKB system. The approach uses full agreement (i.e., exact syntactic match) along with a MaxEnt parse selection model and a statistical dependency parser trained on the same data. The ultimate goal is to develop a hybrid corpus annotation methodology that combines fully automatic annotation and manual parse selection, in order to make the annotation task more efficient while maintaining high accuracy and the high degree of consistency necessary for any foreseen uses of a treebank.
Almost all current dependency parsers classify based on millions of sparse indicator features. Not only do these features generalize poorly, but the cost of feature computation restricts parsing speed significantly. In this work, we propose a novel way of learning a neural network classifier for use in a greedy, transition-based dependency parser. Because this classifier learns and uses just a small number of dense features, it can work very fast, while achieving an about 2% improvement in unlabeled and labeled attachment scores on both English and Chinese datasets. Concretely, our parser is able to parse more than 1000 sentences per second at 92.2% unlabeled attachment score on the English Penn Treebank.
International audience
How does the presence of a categorically related word influence picture naming latencies? In order to test competitive and noncompetitive accounts of lexical selection in spoken word production, we employed the picture-word interference (PWI) paradigm to investigate how conceptual feature overlap influences naming latencies when distractors are category coordinates of the target picture. Mahon et al. (2007. Lexical selection is not by competition: A reinterpretation of semantic interference and facilitation effects in the picture-word interference paradigm. Journal of Experimental Psychology. Learning, Memory, and Cognition, 33(3), 503-535. doi:10.1037/0278-7393.33.3.503 ) reported that semantically close distractors (e.g., zebra) facilitated target picture naming latencies (e.g., HORSE) compared to far distractors (e.g., whale). We failed to replicate a facilitation effect for within-category close versus far target-distractor pairings using near-identical materials based on feature production norms, instead obtaining reliably larger interference effects (Experiments 1 and 2). The interference effect did not show a monotonic increase across multiple levels of within-category semantic distance, although there was evidence of a linear trend when unrelated distractors were included in analyses (Experiment 2). Our results show that semantic interference in PWI is greater for semantically close than for far category coordinate relations, reflecting the extent of conceptual feature overlap between target and distractor. These findings are consistent with the assumptions of prominent competitive lexical selection models of speech production.
English. Assuming the increased need of language resources encoded with shared representation formats, the paper describes a project for the conversion of the multilingual parallel treebank ParTUT in the de facto standard of the Stanford Dependencies (SD) representation. More specifically, it reports the conversion process, currently implemented as a prototype, into the Universal SD format, more oriented to a cross-linguistic perspective and, therefore, more suitable for the purpose of our resource. Italiano. Considerando la crescente necessita di risorse linguistiche codificate in formati ampiamente condivisi, l’articolo presenta un progetto per la conversione di una risorsa multilingue annotata a livello sintattico nel formato, considerato uno standard de facto, delle Stanford Dependencies (SD). Piu precisamente l’articolo descrive il processo di conversione, di cui e attualmente sviluppato un prototipo, nelle Universal Stanford Dependencies, una versione delle SD maggiormente orientata a una prospettiva inter-linguistica e, per questo, particolarmente adatta agli scopi della nostra risorsa.
This article answers the question what is and what is not ellipsis and specifies criteria for identification of elliptical sentences. It reports on an analysis of types of ellipsis from the point of view of semantic representation of sentences. It does not deal with conditions and causes of the constitution of elliptical positions in sentences (when and why is it possible to omit something in a sentence) but it focuses exclusively on the identification of elliptical positions (if there is something omitted and what) and on their semantic representation in a treebank, specifically on their representation on the deep syntactic level of the Prague Dependency Treebanks. The theoretical frame of the approach to ellipsis presented in this article is dependency grammar.
The documentation and analysis of endangered languages is a core component of the linguistic endeavour. Language consultants and linguistic researchers collaborate to generate a variety of data which in turn fuel theoretical discovery and language revitalization. This dissertation describes and evaluates a piece of software designed to facilitate new, and enhance existing, collaboration, documentation, and analysis. But beyond this, it argues for the value of a certain methodological approach to linguistics broadly construed, one in which computation is key and where provisions are made for collaboration, data-sharing and data reuse. The Online Linguistic Database (OLD) is open source software for creating web applications that facilitate collaborative linguistic fieldwork. The OLD allows fieldworkers to continue doing what they are already doing—eliciting, transcribing, recording, and analyzing forms and creating data sets and papers with them—but collaboratively. This point should not be understated: though practises are changing, linguistic fieldwork currently involves a loose network of relatively isolated practitioners and data sets; simply creating the infrastructure for collaboration and data-sharing is half the battle. The other half is creating features and conveniences that make the software worth using. In this domain, the OLD provides automated feedback on lexical consistency of morphological analyses, sophisticated search, the creation and (structural) searching of arbitrarily many corpora and treebanks, and the specification and computational implementation of models of the lexicon, phonology, and morphology, upon which are built practical morphological parsers. The dissertation describes the OLD, motivating its design decisions and arguing that it has the potential to contribute positively to the achievement of the three core goals of linguistic fieldwork, namely documentation, research, and language revitalization. Particular attention is paid to the practical and research-related advantages of the morphophonological modelling capability with examples and evaluations of morphological parsers created for the Blackfoot language.