text stringlengths 4 222k | label int64 0 4 |
|---|---|
Graph-based models (McDonald et al., 2005; McDonald and Pereira, 2006) and transitionbased models (Yamada and Matsumoto, 2003; Nivre and Scholz, 2004) are two dominant paradigms in the dependency parsing community. McDonald and Nivre (2007) have made elaborate analyses about the very different theoretical properties of... | 0 |
Natural language often conveys a sequence of events like "who did what to whom", and extracting structured events from the raw text is a kind of touchstone for machine reading. This is realized by a combination of coreference resolution (called CR, hereafter) and predicate argument structure analysis (called PA, hereaf... | 0 |
Emoji are pictographic characters commonly used in informal written communication, originally in SMS messages, but increasingly in all social media. Emoji represents a range of faces, animals, weather, emotions and activities, compensating for a lack of verbal and non-verbal cues in faceto-face communication (Davis and... | 0 |
The rapid growth of video game industry with new products and technology has significantly increased the popularity of video games. As video games have now become one of the most profitable source of entertainment worldwide, the competition between development companies has increased notably.Catering for gamers' needs ... | 0 |
La déficience auditive est l'un des troubles sensoriels les plus fréquents et touche plus de 6,5% de la population mondiale. Les surdités congénitales représentent une naissance pour mille et dans 90% des cas, l'enfant sourd naît de deux parents normo-entendants (Berland, 2014) . Chez l'enfant, la déficience auditive i... | 0 |
In over a century of practice in descriptive linguistics, the pattern has been to prepare texts and a lexicon to support the construction of a grammar. The grammar includes a description of the phonology and morphosyntax, which inform the representation of the texts and lexicon, in a cyclic arrangement (Crowley, 2007, ... | 0 |
The WMT'16 QE has included the sentence level sub-task. The goal is to predict the amount of effort required to post-edit machine-translated sentences. For this task the organizers provide a parallel corpus of English-German sentences obtained via some machine translation system, as well as corresponding manually post-... | 0 |
Studies about laughter in interaction have been mainly focused on the acoustic or perceptual features, and often observations of the events preceding to it have been the base for claims concerning what laughter is about. (Provine, 1993 ) made a claim that has been subsequently adopted in much of the literature: laughte... | 0 |
The module of morphological analysis and/or synthesis is unavoidable in any language engineering tool for Estonian because of its rich morphology. For example, in information retrieval systems, it is usually desirable to make queries using semantic entities, not using special morphological forms of a word. As the word ... | 0 |
A large portion of the research activity in Natural Language Processing (NLP) is devoted to the development of new algorithms for existing or new tasks. To evaluate the quality of a new method, its performance on unseen datasets is compared to the performance of existing methods. The progress of the field hence crucial... | 0 |
Selecting suitable data is important in all stages of creating an SMT system. For training, the data size plays an essential role, but the data should also be as clean as possible. The new CzEng 1.0 was prepared with the emphasis on data quality and we evaluate it against the previous version to show whether the effect... | 0 |
An Adverse Drug Reaction (ADR) is an injury occurring after a drug (medication) is used at the recommended dosage, for recommended symptoms. This is a area that has already been researched in recent years (Sarker and Gonzalez, 2015; Karimi et al., 2015) , and in which we will contribute with new systems.The proposed sh... | 0 |
In this paper, we introduce a variant of named entity recognition (NER) task, which is event entity recognition (EER). Different from NER task, EER aims at identifying the named entities corresponding to a specific event type. Figure 1 demonstrates some examples of EER. In example a, text contains both the event type '... | 0 |
The Akkadian language was the lingua franca of the Middle East and Egypt in the Late Bronze and Early Iron Ages, spoken or in use from 2500 BCE until its gradual extinction around 100 CE (Oppenheim, 2013) . It was written in cuneiform signswedge-shaped imprints on clay tablets, as depicted in Figure 1 (Walker, 1987) . ... | 0 |
The goal of event extraction is to identify instances of a class of events in text. In addition to identifying the event itself, it also identifies all of the participants and attributes of each event; these are the entities that are involved in that event. The same event might be presented in various expressions, and ... | 0 |
Since the beginning of research on spoken language, many different proposals for the segmentation of spoken language have been proposed. However, there is presently no segmentation system that could be used for large corpora of spoken language, i.e. a system that is linguistically substantiated as well as workable for ... | 0 |
Multiword Expressions (MWEs) are linguistic expressions that consist of multiple lexemes and exhibit some form of idiomaticity. The term idiomaticity denotes the fact that the properties of the whole cannot (fully) be predicted by the properties of the components. Idiomaticity can surface on the lexical, syntactic, sem... | 0 |
Seq2seq neural networks, ever since the successful application in machine translation (Sutskever et al., 2014) , have demonstrated impressive results on dialog generation and spawned a great deal of variants (Vinyals and Le, 2015; Yao et al., 2015; Sordoni et al., 2015; Shang et al., 2015) . The vanilla seq2seq models ... | 0 |
In recent years, with the development of Internet, masses of information provide the majority of Internet users with a lot of convenience. However, with the increase of amount of information, screening redundant information and seeking for the knowledge which users really want from a lot of unstructured texts is gettin... | 0 |
Automatic evaluation of machine translation (MT) systems requires automated procedures to ensure consistency and efficient handling of large amounts of data. In statistical MT systems, automatic evaluation of translations is essential for parameter optimization and system development. Human evaluation is too labor inte... | 0 |
Commonsense has been considered to play a vital role in language understanding (LoBue and Yates, 2011) , and considerable effort has been devoted to building knowledge bases (KBs) that organize commonsense (Zang et al., 2013) . The largest multi-lingual commonsense KB is ConceptNet (Liu and Singh, 2004b) . ConceptNet m... | 0 |
Acoustic cues, generally duration, pausing, and f0, have been demonstrated to be useful for auto- matic segmentation of natural speech (Baron et al., 2002; Hirschberg and Nakatani, 1998; Neiman et al., 1998) . In fact, it is generally accepted that prosodic information is a reliable tool in predicting topic shifts and ... | 0 |
State-of-the art approaches for word alignment are based on probabilistic models. They can be split into joint models (Melamed, 2000; Marcu and Wong, 2002) and conditional models (Brown et al., 1993; Vogel et al., 1996; Wang and Waibel, 1998; Toutanova et al., 2002; Sumita et al., 2004; Deng and Byrne, 2005; Fraser and... | 0 |
End-to-end neural models have shown a great potential in building flexible goal-oriented dialog systems. They can be directly trained on past dialogs without any domain-specific handcrafting, which makes it easy to automatically scale up to new domains . However, these models are normally data-hungry and have only been... | 0 |
One basic task in natural language processing (NLP), as well as other disciplines like computational biology (Waterman, 1995) , is comparing sequences of symbols with each other, deciding about their similarity. In NLP, sequences are designated as sentences, consisting of words.In a first approximation, sentences are c... | 0 |
Automatic Post-Editing (APE) is the task of automatically correcting errors in a given the machine translation (MT) output to generate a better translation (Chatterjee et al., 2018) . Because APE can be regarded as a sequence-to-sequence problem, MT techniques have been previously applied to this task. Subsequently, it... | 0 |
Primary experimental articles (i.e., papers that describe original experimental work) provide the crucial raw material for all other subsequent scientific research. However, the drastically growing number of scientific literature makes it increasingly difficult for domain experts to efficiently utilize them. Automatic ... | 0 |
Software publishers used to rely on documentation sets (manuals) and online support (knowledge base) articles to assist their customers or users with the installation, maintenance or troubleshooting of products. With the advent of Web 2.0 communica-tion channels (such as community forums or social media), users have be... | 0 |
Pre-trained models for vision-and-language tasks have made remarkable progress recently (Lu et al., 2019; Su et al., 2020; . Existing pre-trained models either focus on text-to-image synthesis or image-to-text generation Cho et al., 2021) . These models are often pre-trained with image-text pairs which are aligned in s... | 0 |
A task-oriented spoken dialogue system (SDS) is a system that can continuously interact with a human to accomplish a predefined task through speech. It usually consists of three modules: input, output, and control. The control module is also referred to as dialogue management Yu et al., 2014) . It has two missions: dia... | 0 |
In the past few years there has been a burgeoning interest in syntax-based approaches to statistical machine translation. Most of this effort has been directed at the translation model rather than the language model; however, the level of syntactic divergence in parallel text makes it difficult to reliably learn syntax... | 0 |
Les résultats présentés ici sont la première étape d'un projet dont le but est d'étiqueter et d'analyser des textes volumineux en recourant le moins possible à des connaissances préalables qu'il est nécessaire de spécifier manuellement, particulièrement dans des contextes où des textes étiquetés à la main sont inexista... | 0 |
The development of conversational agents (such as mobile assistants, chatbots or interactive robots) is increasingly based on data-driven methods aiming to infer conversational patterns from dialogue data. One major trend in the last recent years is the emergence of neural conversation models (Vinyals and Le, 2015; Sor... | 0 |
Voice assistants are becoming pervasive in our lives. In order to support their users with different tasks (for example, booking a table at a restaurant for a specific time), they rely on so-called Natural Language Understanding (NLU) components, such as intent classification or slot filling (Tur and De Mori, 2011) . N... | 0 |
The rapid expansion of social media and other online channels has made readable information available at an astounding rate. However, the accessibility of this information is often limited by the complexity of this information, especially among readers with low literacy levels in the language of the text and those with... | 0 |
Over the last few years there has been an increasing number of studies that investigate the possibilities of using literature-specific MT systems in literary translation (see e.g. Lee 2011; Besacier 2014; Toral & Way 2015; Toral & Way 2018) . As stimulating as these studies are, most of them do not discuss any narrativ... | 0 |
Named Entity recognition has been getting much attention in NLP research in recent years, since it is seen as significant component of higher level NLP tasks such as information distillation and question answering. Most successful approaches to NER employ machine learning techniques, which require supervised training d... | 0 |
The micro-blogging platform Twitter is a central venue for online discussions and argumentation. This service has also been widely used to disseminate information during emergencies and natural disasters, and to mobilize support for social and political movements (Lotan, Graeff, Ananny, Gaffney, & Pearce, 2011) . As wi... | 0 |
Interest in applying natural language processing (NLP) technology to medical information has increased in recent years. Much of this work has been focused on information retrieval and extraction from clinical notes, electronic medical records, and biomedical academic literature, but there has been some work in directly... | 0 |
A substantial amount of clinical data is locked away in a non-standardised form of clinical language, which if standardised could be usefully mined to improve processes in the work of clinical wards, and to gain greater understanding of patient care as well as the progression of diseases. However in some clinical conte... | 0 |
In this paper we focus on user utterance understanding, where a conversational system has to interpret the content of a user dialogue turn. At this step, most of conversational systems try to capture both the intent of the utterance and the relevant entities and relations that are mentioned. As an example, given a user... | 0 |
The prompt-based fine-tuning method reduces the gap between pre-training and fine-tuning by forming the fine-tuning task into a masking language problem. A language prompt is a piece of text appended to the query input enabling the model to come up with better predictions (Schick and Schütze, 2021; Tam et al., 2021) . ... | 0 |
Tree Adjoining Grammar (TAG) was introduced by/Joshi et al. 1975/ as a formalism for linguistic description. A TAG's basic component is a finite set of elementary trees, each of which is a domain of locality, and can be viewed as a minimal linguistic structure.A TAG comprises of two kinds of elementary trees: initial t... | 0 |
WordNet (WN), which is a large scale, domain-dependent semantic network of English words, provides a broad-coverage of lexical information. It represents a system of semantic relations among words, between words and synsets' , and between synsets themselves (Miller, 1990 (Miller, , 1995 . Two features of the system are... | 0 |
Unsupervised learning of grammar from text ('grammar induction') is of great theoretical and practical importance. It can shed light on language acquisition by humans and on the general structure of language, and it can potentially assist NLP applications that utilize parser output. The problem has attracted researcher... | 0 |
Spanish (or castellano), a descendant of Latin, is currently the language with the second largest number of native speakers after Mandarin Chinese, namely around 414 million people (Lewis et al., 2014) . Spanish has a large number of regional varieties across Spain and the Americas (Lipski, 1994) . 1 They diverge in sp... | 0 |
With the increasing internet access to Arab populations, their contributions to internet content are growing in a remarkable way. Indeed, the internet penetration rate increased from 30.3% to 51.6% between 2012 and 2019 in Arab countries (ITU 2019) . Additionally, there are multiple social networking platforms facilita... | 0 |
Broad-coverage knowledge graphs such as Freebase, Wikidata, and NELL are increasingly being used in many NLP and AI tasks. For instance, DBpedia and YAGO were vital for IBM's Watson! Jeopardy system (Welty et al., 2012) . Google's Knowledge Graph is tightly integrated into its search engine, yielding improved responses... | 0 |
Dialog systems are concerned with replicating the human ability to make conversation. In a generative dialog system, the model aims at generating coherent and informative responses given a dialog context and, optionally, some external information through knowledge bases (Wen et al., 2017) or annotations e.g. belief sta... | 0 |
Automatic drug name recognition has mostly been studied in terms of extracting drug names from medical documents and biomedical articles (Liu et al., 2015) . However, expanding the same task to extracting drug names from tweets poses a lot more challenges. Tweets are shorter and do not provide enough context compared t... | 0 |
Members of a community rely on a shared language for communication, one which evolves naturally, in situ, and is shaped by the norms and mores of the community (Gumperz, 1968) . Norms promote Entitativity, the perception of group identity (Allan and Burridge, 2006) . Norms may be explicit and codified into law or so su... | 0 |
Machines built to assist humans who engage with texts to seek information ought to be designed with an awareness of the information need. Abstractly, the human's need should define the lens through which the system views the text in order to find desired information. Existing information-seeking machine reading dataset... | 0 |
Un système de recherche documentaire doit trouver dans une base de documents (typiquement une collection de textes) les documents répondantsà une requête posée (en langage naturel) par un utilisateur.En général, un système de recherche documentaire répondraà une requête posée par une liste ordonnée de documents. Le but... | 0 |
Semantic fluency tasks, in which patients undergoing neuropsychological evaluation must list as many items as possible in a particular semantic category in a fixed, brief period of time, are widely used by clinicians to evaluate language, development, and cognition. Performance on such tasks is usually measured in term... | 0 |
The problem of situated natural language generation (NLG)-i.e., of generating natural language in the context of a physical (or virtual) environment-has received increasing attention in the past few years. On the one hand, this is because it is the foundation of various emerging applications, including human-robot inte... | 0 |
Though computer systems aiding retrieval from bibliographic or full-text databases have been available for more than two decades, it is only in recent years that many people are becoming concerned about the serious limitations of those systems regarding effectiveness in finding desired references (Blair and Maron 1985)... | 0 |
In error analysis, we often blame morphology (Nivre, 2007; Bender, 2009) , i.e., the productive inflection and derivation of new word forms. Morphology has been argued to be a major source of error in syntactic parsing , semantic parsing (Şahin and Steedman, 2018), machine translation (Irvine et al., 2013; Burlot and Y... | 0 |
CHIMERA (Bojar et al., 2013; Tamchyna et al., 2014) is our English-to-Czech MT system designed as a combination of three very different components:• TectoMT (Popel andŽabokrtský, 2010) , a deep-syntactic transfer-based system,• Moses (Koehn et al., 2007) , where we use a factored phrase-based setup with large language ... | 0 |
In Switzerland, two thirds of the population speak Swiss German, which is primarily a spoken language with many regional dialects. Swiss German has no standardised written form (Honnet et al., 2018) , thus written communication relies on Standard German. Swiss German is widely used on Swiss TV, for example in news repo... | 0 |
Stitching portions of text together into a sentence is a crucial first step in abstractive summarization. It involves choosing which sentences to fuse, what content from each of them to retain and how best to present that information (Elsner and Santhanam, 2011) . A major challenge in fusing sentences is to establish c... | 0 |
In trying to solve the problem of building programs that interact in natural language no great attention has been paid to formalisms for the structural analysis of dialogues. However, such formalisms may support a clear description of the ni L tory of the whole conversation, and convey reasoning power when associated t... | 0 |
The Prague Dependency Treebank 2.0 (PDT 2.0; Hajič et al., 2006 ) is a manually annotated corpus of Czech. It belongs to the most complex end elaborate linguistically annotated treebanks in the world. The texts are annotated on three layers of language description: morphological, analytical (which expresses the surface... | 0 |
POS tagging is a well-explored problem in NLP, and highly-accurate taggers have been built using statistical and probabilistic methods for decades. These taggers are typically trained using already-annotated text, from which the probabilities of POS tags being appropriate for certain word tokens are calculated based on... | 0 |
While black box models like GPT-3 (Brown et al., 2020) demonstrate impressive performance on a variety of isolated benchmarks, they are still subject to significant drawbacks (Marcus, 2018) . In particular, as AI systems reach wider adoption, explainability and interpretability become critical features. It is our belie... | 0 |
In 2019, 93% of Americans identify incivility as a problem, with 68% classifying it as a "major" problem, and those who experienced incivility faced on average 10.2 uncivil interactions each week (Weber Shandwick et al., 2019) . Of those who expect civility to get worse, "social media/the Internet" tops the list of wha... | 0 |
The wide spread of fake news on online social media has influenced public trust (Knight Foundation, 2018) and poses real-world threats on politics (Fisher et al., 2016) , finance (ElBoghdady, 2013) , public health (Naeem and Bhatti, 2020), etc. Under such severe circumstances, automatically detecting fake news has been... | 0 |
In medicine, examination of a patient's diseases is described in many reports in multiple specialties. Each report specializes in a specific aspect of the diseases, such as the chest, head and bones. For precise treatments, understanding the holistic picture of the patient's clinical history is critical. Unstructured t... | 0 |
Consider the following example: Example 1 John made a handbag from an inner-tube. a. He sold it for twenty dollars. b. *He sold them for fifty dollars. c. He had taken it from his brother's car. d. Neither of them was particularly useful.Here two entities are introduced via indefinite noun phrases (NPs) in the first se... | 0 |
This is a programmatic paper (and work in progress) motivated by the challenge of automatically identifying and representing contradictions in medical guidelines. In this paper we take the perspective of building a natural language understanding system that can properly represent disagreements. On the practical side, w... | 0 |
Beam-search has been applied to transition-based dependency parsing in recent studies (Zhang and Clark, 2008; Huang and Sagae, 2010; Hatori et al., 2011) . In addition to reducing search errors compared to greedy search, it also enables the use of global models that accommodate richer non-local features without overfit... | 0 |
This paper presents a prototype shallow-transfer rulebased machine translation system between Kazakh and Tatar. The paper will be laid out as follows: Section 2 gives a short review of some previous work in the area of Turkic-Turkic language machine translation; Section 3 introduces Kazakh and Tatar and compares their ... | 0 |
Distributional semantic methods that use aggregate syntactic and textual word co-occurrence behaviour have been applied successfully to many natural language processing tasks. These methods are often deployed as part of a pipeline to solve an engineering problem, and evaluated by classification or correlation performan... | 0 |
Due to Twitter's massive popularity, it has become a tool used by first responders-those who provide first-hand aid in times of crisis-to understand crisis situations and identify the people in the most dire need of assistance (United Nations, 2012) . To do this, first responders can survey "geotagged" tweets: those wh... | 0 |
Biomedical experts (e.g., researchers, clinical doctors) routinely need to search the biomedical literature to support research hypotheses, treat rare syndromes, follow best practices etc. The most widely used biomedical search engine is PUBMED, with more than 24 million biomedical references and abstracts, mostly of j... | 0 |
Text generation spans a broad set of natural language processing tasks that aim at generating natural language from input data or machine representations. Such tasks include machine translation (Bahdanau et al., 2014; Brown et al., 1990) , dialog systems (Williams and Young, 2007; Serban et al., 2016) , text summarizat... | 0 |
The task of definition modeling, introduced by Noraset et al. (2017) , consists in generating the dictionary definition of a specific word: for instance, given the word "monotreme" as input, the system would need to produce a definition such as "any of an order (Monotremata) of egg-laying mammals comprising the platypu... | 0 |
Due to the increasing number of unstructured documents available on the Web and from other sources, developing techniques that automatically extract knowledge from the documents has been of paramount importance. Among many aspects of extracting knowledge from documents, the extraction of temporal information is recentl... | 0 |
Task-oriented dialogue systems enable users to accomplish tasks, such as ticket booking, restaurant reservation, and customer support, by interacting in natural language. The ability to accurately track the user's requirements during the dialogue is crucial to enable a consistent and effective dialogue . Dialogue syste... | 0 |
Weighted finite-state machines (WFSMs) have a storied role in NLP. They are a useful formalism for speech recognition (Mohri et al., 2002) , machine transliteration (Knight and Graehl, 1998) , morphology (Geyken and Hanneforth, 2005; Lindén et al., 2009) and phonology (Cotterell et al., 2015) inter alia. Indeed, WFSMs ... | 0 |
In this paper we present a detailed scheme for annotating expressions of opinions, beliefs, emotions, sentiment, speculation and other private states in newspaper articles. Private state is a general term that covers mental and emotional states, which cannot be directly observed or verified (Quirk et al., 1985) . For e... | 0 |
With the popularity of social media like Twitter, offensive language has become a serious problem (Zampieri et al., 2019b ) on these media platforms. People have to face with abusive behavior from others in social media from time to time. To solve this problem, finding a method to identify and categorize offensive lang... | 0 |
The popularity of the Internet and the unprecedented reach of social media platforms allow users to express their opinions on a wide range of topics. Thereby, Sentiment Analysis (SA) has become a cornerstone for many applications such as digital marketing, product review analysis, customer feedback, social media monito... | 0 |
Accompanying the revolution of pre-trained models in many NLP applications, such as sentiment analysis (Xu et al., 2019a) , question answering (Yang et al., 2019b) , information retrieval (Yang et al., 2019c) , and text generation (Raffel et al., 2020) , many related technologies have been deployed on the cloud to proc... | 0 |
In 1997, Mark Seligman wrote that "the Internet offers a tremendous opportunity for experiments with real-time machine translation (MT) of dialogues" (Seligman, 1997) . In December of the same year, SYSTRAN and AltaVista launched "the first widely available, real-time, high-speed and free translation service on the Int... | 0 |
Although superlative noun phrases (the nation's largest milk producer, the most complex armscontrol talks ever attempted, etc.) received considerable attention in formal linguistics (Szabolcsi, 1986; Gawron, 1995; Heim, 1999; Farkas and Kiss, 2000) , this interest is not mirrored in computational linguistics and NLP. O... | 0 |
Information diffusion is a process of spreading information or content within a network via a particular path or pattern. A significant amount of research has been done in this area in the past few years. However, most of the previous efforts considered only network topology for the diffusion process.To understand the ... | 0 |
It is widely recognized that Japanese manner of motion verbs do not tolerate a so-called GOAL expression -ni, as observed in (1) (see e.g. Yoneyama (1986) , Kageyama and Yumoto (1997) , Ueno and Kageyama (2000) ). 1 ' 2 The same is true of Korean (cf. Lee (1999) Taro-TOP station-GOAL run-PAST `(Lit.)Taro ran to the sta... | 0 |
Natural Language Understanding (NLU) models are used for different tasks such as questionanswering (Hirschman and Gaizauskas, 2001) , machine translation (Macherey et al., 2001 ) and text summarization (Tas and Kiyani, 2007) . These models are often trained on crowd-sourced data that may contain sensitive information s... | 0 |
The work reported here grew out of our attempt to improve on the o (n 2) performance of the SIMD parallel parser described in (Thompson 1991) . Rather than start with a commitment to a specific SIMD architecture, as that work had, we agreed that the best place to start was with a more abstract architecture-independent ... | 0 |
The reason of building this corpus is that having a bilingual semantic corpus with refined semantic role information is expected to bring significant benefit to the task of Statistical Machine Translation (SMT). Since the semantic constituent is less variant during translation in comparison with the syntactic constitue... | 0 |
Everything has two sides, and information technology is no exception. On the one hand, the rapid development of information technology makes it easy for people to communicate and discuss on social platforms Mahesan, 2019, 2020a,b) . On the other hand, the rapid development of information technology has also led to the ... | 0 |
Computational tools for automatically detecting and categorizing abusive online content are now widely used for content moderation, to enforce and monitor regulatory and legal standards, and to study the dynamics of online abuse (Williams, 2019; Vidgen et al., 2019; Fortuna and Nunes, 2018) . These tools enable abusive... | 0 |
Current models of memory (Marr, 1971; Anderson et al., 1977; Murdock, 1982; McClelland et al., 1995; Howard and Kahana, 2002) involve a continuous activation-based (or 'working') memory, typically modeled as a vector representing the current firing pattern of neurons or neural clusters in the cortex. This activation-ba... | 0 |
Les systèmes de traduction statistiques actuels intègrent différents modèles qui mettent en jeu, lors du décodage, le plus d'informations disponibles afin de produire les meilleures traductions possibles. En particulier, dans leur version standard, ces systèmes embarquent un modèle de traduction qui probabilise la corr... | 0 |
The enormous quantity of self-reporting about medication effects in social media provides an opportunity for additional surveillance of medication effects, supplementing traditional health-related reporting systems (Edo-Osagie et al., 2020) . Task 2 of the SMM4H Shared Task challenge (Klein et al., 2020) was defined as... | 0 |
Most work on the computational analysis of sentiment and perspective relies on lexical features. This makes sense, since an author's choice of words is often used to express overt opinions (e.g. describing healthcare reform as idiotic or wonderful) or to frame a discussion in order to convey a perspective more implicit... | 0 |
Vocal activity detection (VAD) is an important technology for any application with an automatic speech recognition (ASR) front end. In meetings, participants typically vocalize for only a fraction of the recorded time. Their temporally contiguous contributions should be identified prior to ASR in order to leverage spea... | 0 |
When developing English→German speech translation systems for the IWSLT 2020 evaluation, we had the following goals:• To obtain the best possible translation quality with the baseline cascaded approach. This includes data filtering, weighting, and domain adaptation for the MT component, hybrid ASR (Section 2.1) with a ... | 0 |
Since Montague (1970) , formal compositional semantics has provided successful accounts of linguistic phenomena using logical expressions along with syntactic structures. In recent years, with the development of wide-coverage parsers such as Combinatory Categorial Grammar (CCG) parsers (Clark and Curran, 2007) , some o... | 0 |
Web logging (blogging) and its social impact have recently attracted considerable public and scientific interest. One use of blogs is as a community discussion forum, especially for political discussion and debate. Blogging has arguably opened a new channel for huge numbers of people to express their views with unprece... | 0 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.