Ambiguous Noun Research Articles

The ambiguous nature of complex semantic types: an experimental investigation

Abstract Words with complex semantic types such as book are characterised by a multiplicity of interpretations that are not mutually exclusive (e.g., as a physical object and/or informational content). Their status with respect to lexical ambiguity is notoriously unclear, and it is debatable whether complex types are a particular form of polysemy (closely related to metonymy) or whether they belong to monosemy. In this study, we investigate the nature of complex types by conducting two experiments on ambiguous nouns in French. The first experiment collects speakers’ judgements about the sameness of meaning between different uses of complex-type, metonymic and monosemous words. The second experiment uses a priming paradigm and a sensicality task to investigate the online processing of complex-type words, as opposed to metonymic and monosemous words. Overall results indicate that, on a continuum of lexical ambiguity, complex types are closer to monosemy than to metonymy. The different interpretations of complex-type words are highly connected and fall under the same meaning, arguably in relation to a unique reference. These results suggest that complex types are associated with single underspecified entries in the mental lexicon. Moreover, they highlight the need for a model of lexical representations of ambiguous words that can account for the difference between complex types and metonymy.

The role of category ambiguity in normal and impaired lexical processing: can you paint without the paint?

Many words are categorially ambiguous and can be used as a verb (to paint) or as a noun (the paint) due to the presence of unpronounced morphology or "zero morphology". On this account, the verb "paint" is derived from the noun "paint" through the addition of a silent category-changing morpheme. Past studies have uncovered the syntactic and semantic properties of these categorially ambiguous words, but no research has been conducted on how people process them during normal or impaired lexical processing. Are these two different uses of "paint" processed in the same way? Does this morphosyntactic structure have an effect on online sentence processing? This study presents two experiments that investigate the effect of morphosyntactic complexity in categorially ambiguous words presented in isolation (experiment 1) and in a sentential context (experiment 2). The first experiment tested the ability to process categorially unambiguous and ambiguous nouns and verbs in 30 healthy older adults and 12 individuals with aphasia, using a forced choice phrasal-completion task, in which individuals choose whether the or to is most compatible with target words. Healthy controls and individuals with fluent aphasia all showed: (1) a bias toward the base category in selection rates for the and to, where the was selected more frequently for words identified to be base nouns, and to was selected more frequently for base verbs, and (2) longer reaction times for ambiguous (over unambiguous) words. However, individuals with non-fluent agrammatic aphasia showed a base-category effect only for nouns, with chance performance for verbs. The second experiment, using an eye-tracking while reading paradigm with 56 young healthy adults, showed a reading time slowdown for derived forms (to paint) compared to their base-category counterparts (the paint) in sentence contexts. These findings suggest that categorially ambiguous words likely share a common root, and are related by zero-derivation, and that impaired access to the base-category (i.e., verbs like to visit) precludes associated morphological processes and therefore the retrieval of the derived-category (i.e., nouns like the visit) in non-fluent agrammatic aphasia. This study provides insights into the theory of zero morphology, and the principles that need to be accounted for in models of the lexicon.

When morphology meets regular polysemy

This study investigates the existence of complex morphological patterns that include a regular polysemy extension. Such complex patterns derive ambiguous words whose semantic types are both related and specified. To address this issue, nouns derived from verbs in French are investigated. A sample of 2,953 deverbal nouns ending with 46 suffixes are collected and systematically analyzed with respect to semantic properties and lexical ambiguity. Thirty-eight complex patterns are identified. The formation of ambiguous nouns whose meanings fit into a regular polysemy pattern appears as a discriminative property of some suffixes. Different kinds of patterns are further distinguished, depending on whether secondary meanings in complex patterns can be directly related to a base verb or not. In accordance with the literature on the polysemy of nominalizations, the semantic categories of EVENT and RESULT are prominent in morphological patterns, but other semantic types such as AGENT, INSTITUTION and COGNITIVE OBJECT are importantly represented as well. In addition, metonymy is the main semantic figure observed in complex patterns. Overall, the article sheds new light on the relationship between word formation and lexical polysemy by describing the possibility of direct association between them.

Suppression of non-selected solutions as a possible brain mechanism for ambiguity resolution in the word fragment task completion task

Brain systems dealing with multiple meanings of ambiguous stimuli are relatively well studied, while the processing of non-selected meanings is less investigated in the neurophysiological literature and provokes controversy between existing theories. It is debated whether these meanings are actively suppressed and, if yes, whether suppression characterizes any task that involves alternative solutions or only those tasks that emphasize semantic processing or the existence of alternatives. The current functional MRI event-related study used a modified version of the word fragment completion task to reveal brain mechanisms involved in implicit processing of the non-selected solutions of ambiguous fragments. The stimuli were pairs of fragmented adjectives and nouns. Noun fragments could have one or two solutions (resulting in two words with unrelated meanings). Adjective fragments had one solution and created contexts strongly suggesting one solution for ambiguous noun fragments. All fragmented nouns were presented twice during the experiment (with two different adjectives). We revealed that ambiguity resolution was associated with a reduced BOLD signal within several regions related to language processing, including the anterior hippocampi and amygdala and posterior lateral temporal cortex. Obtained findings were interpreted as resulting from brain activity inhibition, which underlies a hypothesized mechanism of suppression of non-selected solutions.

Hierarchy in language interpretation: evidence from behavioural experiments and computational modelling

ABSTRACT It has long been recognised that phrases and sentences are organised hierarchically, but many computational models of language treat them as sequences of words without computing constituent structure. Against this background, we conducted two experiments which showed that participants interpret ambiguous noun phrases, such as second blue ball, in terms of their abstract hierarchical structure rather than their linear surface order. When a neural network model was tested on this task, it could simulate such “hierarchical” behaviour. However, when we changed the training data such that they were not entirely unambiguous anymore, the model stopped generalising in a human-like way. It did not systematically generalise to novel items, and when it was trained on ambiguous trials, it strongly favoured the linear interpretation. We argue that these models should be endowed with a bias to make generalisations over hierarchical structure in order to be cognitively adequate models of human language.

Distributional Models of Category Concepts Based on Names of Category Members.

Cognitive scientists have long used distributional semantic representations of categories. The predominant approach uses distributional representations of category-denoting nouns, such as "city" for the category city. We propose a novel scheme that represents categories as prototypes over representations of names of its members, such as "Barcelona," "Mumbai," and "Wuhan" for the category city. This name-based representation empirically outperforms the noun-based representation on two experiments (modeling human judgments of category relatedness and predicting category membership) with particular improvements for ambiguous nouns. We discuss the model complexity of both classes of models and argue that the name-based model has superior explanatory potential with regard to conceptacquisition.

Reanalysis processes in non-native sentence comprehension

AbstractWe report two offline and two eye-movement experiments examining non-native (L2) sentence processing during and after reanalysis of temporarily ambiguous sentences like “While Mary dressed the baby laughed happily”. Such sentences cause reanalysis at the main clause verb (“laughed”), as the temporarily ambiguous noun phrase (“the baby”) may initially be misanalysed as the direct object of the subordinate clause verb (“dressed”). The offline experiments revealed that L2ers have difficulty reanalysing temporarily ambiguous sentences with a greater persistence of the initially assigned misinterpretation than native (L1) speakers. In the eye-movement experiments, we found that L2ers complete reanalysis similarly to L1ers but fail to fully erase the memory trace of the initially assigned interpretation. Our results suggested that the source of L2 reanalysis difficulty is a failure to erase the initially assigned misinterpretation from memory rather than a failure to conduct syntactic reanalysis.

Aportaciones a la historia del género ambiguo en español: a propósito del sustantivo deverbal tizne

The New Grammar of the Spanish Language (2009) defines ambiguous nouns in terms of gender as those that refer to the same inanimate entity, both male and female, «but do not usually designate sexed beings». Some can be analyzed as late developments, after the restructuring of gender that followed the loss of neutral and casual flexion, together with verbs roots that show similar vacillations to those originated after the evolutionary processes that emerged in the transition from Latin into Romance languages. Among them we find the case of the word tizne, whose variations throughout the history of the Spanish language will be the fundamental object of the present study.

All-words Word Sense Disambiguation for Russian Using Automatically Generated Text Collection

Abstract The limited amount of the sense annotated data is a big challenge for the word sense disambiguation task. As a solution to this problem, we propose an algorithm of automatic generation and labelling of the training collections based on the monosemous relatives concept. In this article we explore the limits of this algorithm: we employ it to harvest training collections for all ambiguous nouns, verbs and adjectives presented in RuWordNet thesaurus and then evaluate the quality of the obtained collections. We demonstrate that our approach can create high-quality labelled collections with almost full-coverage of the RuWordNet polysemous words. Furthermore, we show that our method can be applied to the Word-in-Context task.

Italian Grammatical Gender: A Corpus Study from a Second Language Acquisition Perspective

Abstract The three objectives of this corpus study are first to describe how the gender of inanimate nouns is morphologically expressed in Italian, then determine the extent and reliability of contextual cues to Italian gender by analyzing 5,038 contextualized determiner phrases (DPs) drawn from 40 current newspaper and magazine articles and finally, consider learnability implications for Anglophone learners of Italian as a second language. A detailed descriptive account of gender morphology characterized by vocalic endings as portmanteau morphemes reveals several idiosyncrasies and complexities while the results of the written corpus showing that 33.4% of ambiguous feminine nouns and 31.0% of ambiguous masculine nouns are not gender-marked at all are argued to present significant difficulties for second language learners particularly for those who have a low tolerance for ambiguity.

Gender agreement hierarchy in common gender and epicene nouns in Spanish

Gender is a grammatical category defined as an abstract morphosyntactic feature of nouns reflected in characteristics of associated words (i.e. agreement) (Hockett, 1958; Corbett, 1991). Agreement is, in fact, easily established in “transparent” nouns which follow either semantic or formal rule of gender agreement. However, when we deal with ambiguous nouns regarding their gender, agreement is not straightforward. In this article we aim to pursue two main goals. Firstly, to review and briefly describe grammatical gender system in Spanish (§1) with a special focus on so called “ambiguous” or “problematic” nouns (§2). Secondly, to review agreement hierarchy theories and explore if they are applicable for Spanish epicenes and common gender nouns (§3). Discussion and conclusion remarks are presented in (§4).

Finding identity in the midst of ambiguity: case and number disambiguation in Basque

ABSTRACT Restrictive contextual information has been found to bias syntactic disambiguation, when only one alternative leads to a meaningful interpretation. The current study tests whether disambiguation can be influenced by nonrestrictive cues – when several alternatives are equally plausible. We first evaluated if modifier number biased the disambiguation of number- and case-ambiguous nouns in Basque. In a noun phrase comprehension paradigm, ambiguous noun number judgments were biased by preceding modifier number. Then, using a preamble completion paradigm, we examined whether headnoun disambiguation and thus sentence completion was also biased by modifier number. Our results suggest that nonrestrictive information (singular and plural number) can affect disambiguation. We also report task differences in the overall interpretation of ambiguous Basque nouns, as well modifier-induced agreement errors. We suggest that the parser uses any available context information when there is ambiguity, including preceding modifier markings.

Comprehension of the copula: preschoolers (and sometimes adults) ignore subject-verb agreement during sentence processing.

Subject-verb (SV) agreement helps listeners interpret the number condition of ambiguous nouns (The sheep is/are fat), yet it remains unclear whether young children use agreement to comprehend newly encountered nouns. Preschoolers and adults completed a forced choice task where sentences contained singular vs. plural copulas (Where is/are the [novel noun(s)]?). Novel nouns were either morphologically unambiguous (tup/tups) or ambiguous (/geks/ = singular: gex / plural: gecks). Preschoolers (and some adults) ignored the singular copula, interpreting /ks/-final words as plural, raising questions about the role of SV agreement in learners' sentence comprehension and the status of is in Australian English.

Distinguishing Old From New Referents During Discourse Comprehension: Evidence From ERPs and Oscillations.

In this EEG study, we used pre-registered and exploratory ERP and time-frequency analyses to investigate the resolution of anaphoric and non-anaphoric noun phrases during discourse comprehension. Participants listened to story contexts that described two antecedents, and subsequently read a target sentence with a critical noun phrase that lexically matched one antecedent (‘old’), matched two antecedents (‘ambiguous’), partially matched one antecedent in terms of semantic features (‘partial-match’), or introduced another referent (non-anaphoric, ‘new’). After each target sentence, participants judged whether the noun referred back to an antecedent (i.e., an ‘old/new’ judgment), which was easiest for ambiguous nouns and hardest for partially matching nouns. The noun-elicited N400 ERP component demonstrated initial sensitivity to repetition and semantic overlap, corresponding to repetition and semantic priming effects, respectively. New and partially matching nouns both elicited a subsequent frontal positivity, which suggested that partially matching anaphors may have been processed as new nouns temporarily. ERPs in an even later time window and ERPs time-locked to sentence-final words suggested that new and partially matching nouns had different effects on comprehension, with partially matching nouns incurring additional processing costs up to the end of the sentence. In contrast to the ERP results, the time-frequency results primarily demonstrated sensitivity to noun repetition, and did not differentiate partially matching anaphors from new nouns. In sum, our results show the ERP and time-frequency effects of referent repetition during discourse comprehension, and demonstrate the potentially demanding nature of establishing the anaphoric meaning of a novel noun.

Word sense disambiguation for Punjabi language using deep learning techniques

Word sense disambiguation (WSD) identifies the right meaning of the word in the given context. It is an indispensable and critical application for all the natural language processing tasks. In this paper, two deep learning techniques multilayer perceptron and long short-term memory (LSTM) have been individually inspected on the word vectors of 66 ambiguous Punjabi nouns for an explicit WSD system of Punjabi language. The inputs to the deep learning techniques are the simple word vectors derived directly from manually sense-tagged corpus of Punjabi language. The multilayer perceptron has outperformed the LSTM deep learning technique for WSD task of Punjabi language. Six traditional supervised machine learning techniques have also been tested on same dataset using unigram and bigram feature sets. A comparison between deep learning techniques and traditional six supervised machine learning techniques clearly indicates that the deep learning techniques using simple word vectors outperforms the earlier techniques.

Sense disambiguation for Punjabi language using supervised machine learning techniques

Automatic identification of a meaning of a word in a context is termed as Word Sense Disambiguation (WSD). It is a vital and hard artificial intelligence problem used in several natural language processing applications like machine translation, question answering, information retrieval, etc. In this paper, an explicit WSD system for Punjabi language using supervised techniques has been analysed. The sense tagged corpus of 150 ambiguous Punjabi noun words has been manually prepared. The six supervised machine learning techniques Decision List, Decision Tree, Naive Bayes, K-Nearest Neighbour (K-NN), Random Forest and Support Vector Machines (SVM) have been investigated in this proposed work. Every classifier has used same feature space encompassing lexical (unigram, bigram, collocations, and co-occurrence) and syntactic (part of speech) count based features. The semantic features of Punjabi language have been devised from the unlabelled Punjabi Wikipedia text using word2vec continuous bag of word and skip gram shallow neural network models. Two deep learning neural network classifiers multilayer perceptron and long short term memory have also been applied for WSD of Punjabi words. The word embedding features have experimented on six classifiers for the Punjabi WSD task. It has been observed that the performance of the supervised classifiers applied for the WSD task of Punjabi language has been enhanced with the application of word embedding features. In this work, an accuracy of 84% has been achieved by LSTM classifier using word embedding feature.

Integrating Weakly Supervised Word Sense Disambiguation into Neural Machine Translation

This paper demonstrates that word sense disambiguation (WSD) can improve neural machine translation (NMT) by widening the source context considered when modeling the senses of potentially ambiguous words. We first introduce three adaptive clustering algorithms for WSD, based on k-means, Chinese restaurant processes, and random walks, which are then applied to large word contexts represented in a low-rank space and evaluated on SemEval shared-task data. We then learn word vectors jointly with sense vectors defined by our best WSD method, within a state-of-the-art NMT system. We show that the concatenation of these vectors, and the use of a sense selection mechanism based on the weighted average of sense vectors, outperforms several baselines including sense-aware ones. This is demonstrated by translation on five language pairs. The improvements are more than 1 BLEU point over strong NMT baselines, +4% accuracy over all ambiguous nouns and verbs, or +20% when scored manually over several challenging words.

Integrating Weakly Supervised Word Sense Disambiguation into Neural Machine Translation

This paper demonstrates that word sense disambiguation (WSD) can improve neural machine translation (NMT) by widening the source context considered when modeling the senses of potentially ambiguous words. We first introduce three adaptive clustering algorithms for WSD, based on k-means, Chinese restaurant processes, and random walks, which are then applied to large word contexts represented in a low-rank space and evaluated on SemEval shared-task data. We then learn word vectors jointly with sense vectors defined by our best WSD method, within a state-of-the-art NMT system. We show that the concatenation of these vectors, and the use of a sense selection mechanism based on the weighted average of sense vectors, outperforms several baselines including sense-aware ones. This is demonstrated by translation on five language pairs. The improvements are above one BLEU point over strong NMT baselines, +4% accuracy over all ambiguous nouns and verbs, or +20% when scored manually over several challenging words.

NAIVE BAYES CLASSIFIER FOR WORD SENSE DISAMBIGUATION OF PUNJABI LANGUAGE

Word Sense Disambiguation (WSD) is the process of identifying the correct sense of the word in the context. The most leading scheme used by WSD is machine learning approach, where a human expert provides examples of correctly disambiguated words, and a machine learning algorithm is used to induce a model from these examples. In this paper, Naive Bayes supervised classifier has been used to disambiguate words of Punjabi language. The feature extraction process plays a vital role in building the supervised machine learning models. For the proposed Punjabi WSD system, Bag of Words (BoW) and collocation models are used separately to extract relevant features. BoW model has used all words around target word while collocation model has used two words before and two words after the target word as features. Both the models have used a common training data set to build the model. It has been observed that the selection of smoothing parameter for Naive Bayes has a significant impact on its performance. This proposed work has been tested on 150 most ambiguous noun words selected form Punjabi WordNet having 6 or more senses. During the process of building the model, fine senses of ambiguous words have been merged to produce coarse sense on the basis of manual analysis of lexical relations of WordNet. The accuracy of the proposed system has been calculated independently for BoW and collocation model. The proposed WSD system achieves an accuracy of 89% for BoW model and 81% for collocation model. It has been concluded that BoW model performs better than the collocation model for WSD task for Punjabi language.

EFL Learners’ Knowledge and Use of Gender Stereotypes: Evidence from Arabic Native Speakers

<p><em>The research explored EFL learners’ knowledge and use of gender stereotypes of common English nouns (e.g., doctor and nurse). In the study, we compared how EFL learners living in Saudi Arabia and native English speakers rated 24 nouns that can be used to refer to either males or females and how they interpreted sentences containing the gender-specific pronouns his and her preceded by one of the three types of these nouns (i.e., male stereotyped, female stereotyped, or gender neutral). The results showed that performance for EFL learners differed from native speakers’ in both tasks. EFL learners rated nouns as generally referring to males more often than did native English speakers. EFL learners were also significantly less likely to interpret her and his as referring to the preceding noun than were native English speakers. The results suggest that in EFL courses, learners are likely to benefit from explicit coverage of gender ambiguous English nouns and the topic of gender stereotyping as an important aspect of vocabulary knowledge.</em></p>