Meanings Of Ambiguous Words Research Articles

Comparison of Pre-trained vs Custom-trained Word Embedding Models for Word Sense Disambiguation

The prime objective of word sense disambiguation (WSD) is to develop such machines that can automatically recognize the actual meaning (sense) of ambiguous words in a sentence. WSD can improve various NLP and HCI challenges. Researchers explored a wide variety of methods to resolve this issue of sense ambiguity. However, majorly, their focus was on English and some other well-reputed languages. Urdu with more than 300 million users and a large amount of electronic text available on the web is still unexplored. In recent years, for a variety of Natural Language Processing tasks, word embedding methods have proven extremely successful. This study evaluates, compares, and applies a variety of word embedding approaches to Urdu Word embedding (both Lexical Sample and All-Words), including pre-trained (Word2Vec, Glove, and FastText) as well as custom-trained (Word2Vec, Glove, and FastText trained on the Ur-Mono corpus). Two benchmark corpora are used for the evaluation in this study: (1) the UAW-WSD-18 corpus and (2) the ULS-WSD-18 corpus. For Urdu All-Words WSD tasks, top results have been achieved (Accuracy=60.07 and F1=0.45) using pre-trained FastText. For the Lexical Sample, WSD has been achieved (Accuracy=70.93 and F1=0.60) using custom-trained GloVe word embedding method.

The Effects of Word Processing, Recognition and Context in Lexical Ambiguity Resolution

This article aims to discuss the effects of word processing, recognition and context in lexical ambiguity resolution, trying to find an answer to how the reader-listener determines the contextually appropriate meaning of a word. Ambiguity resolution is analyzed and explored in two perspectives: the context in which the lexical items appear and the activation of all the meanings that an ambiguous word has. There is no clear-cut answer to lexical ambiguity resolution and there is a great debate about the role of the context in the activation of the meaning of ambiguous words.

Causal Contributions of the Domain-General (Multiple Demand) and the Language-Selective Brain Networks to Perceptual and Semantic Challenges in Speech Comprehension.

Listening to spoken language engages domain-general multiple demand (MD; frontoparietal) regions of the human brain, in addition to domain-selective (frontotemporal) language regions, particularly when comprehension is challenging. However, there is limited evidence that the MD network makes a functional contribution to core aspects of understanding language. In a behavioural study of volunteers (n = 19) with chronic brain lesions, but without aphasia, we assessed the causal role of these networks in perceiving, comprehending, and adapting to spoken sentences made more challenging by acoustic-degradation or lexico-semantic ambiguity. We measured perception of and adaptation to acoustically degraded (noise-vocoded) sentences with a word report task before and after training. Participants with greater damage to MD but not language regions required more vocoder channels to achieve 50% word report, indicating impaired perception. Perception improved following training, reflecting adaptation to acoustic degradation, but adaptation was unrelated to lesion location or extent. Comprehension of spoken sentences with semantically ambiguous words was measured with a sentence coherence judgement task. Accuracy was high and unaffected by lesion location or extent. Adaptation to semantic ambiguity was measured in a subsequent word association task, which showed that availability of lower-frequency meanings of ambiguous words increased following their comprehension (word-meaning priming). Word-meaning priming was reduced for participants with greater damage to language but not MD regions. Language and MD networks make dissociable contributions to challenging speech comprehension: Using recent experience to update word meaning preferences depends on language-selective regions, whereas the domain-general MD network plays a causal role in reporting words from degraded speech.

Homonymy Analysis in Taylor Swift’s Songs

This research aims to determine the types of ambiguous words and the meanings of ambiguous words in Taylor Swift's songs. The researcher uses Ullmann's theory to determine the type of ambiguous word. The research design is a qualitative research. The research instrument is the researcher himself. The data sources for this study are the audio and lyrics of Taylor Swift's songs The data of this study are ambiguous words contained in Taylor Swift's songs. In collecting the data, the researcher applies the non-participant observation method and the note taking technique as a follow-up technique. Data analysis conducts the referential method with the language references as the determinant. The researcher finds out homonyms in the songs whose meanings were not related and the actual meanings matched the context in the song lyrics. This research also focuses on homograph and homophone.

Learning about the meanings of ambiguous words: evidence from a word-meaning priming paradigm with short narratives.

Fluent language comprehension requires people to rapidly activate and integrate context-appropriate word meanings. This process is challenging for meanings of ambiguous words that are comparatively lower in frequency (e.g., the “bird” meaning of “crane”). Priming experiments have shown that recent experience makes such subordinate (less frequent) word meanings more readily available at the next encounter. These experiments used lists of unconnected sentences in which each ambiguity was disambiguated locally by neighbouring words. In natural language, however, disambiguation may occur via more distant contextual cues, embedded in longer, connected communicative contexts. In the present experiment, participants (N = 51) listened to 3-sentence narratives that ended in an ambiguous prime. Cues to disambiguation were relatively distant from the prime; the first sentence of each narrative established a situational context congruent with the subordinate meaning of the prime, but the remainder of the narrative did not provide disambiguating information. Following a short delay, primed subordinate meanings were more readily available (compared with an unprimed control), as assessed by responses in a word association task related to the primed meaning. This work confirms that listeners reliably disambiguate spoken ambiguous words on the basis of cues from wider narrative contexts, and that they retain information about the outcome of these disambiguation processes to inform subsequent encounters of the same word form.

Word Sense Induction with Attentive Context Clustering

This paper presents ACCWSI (Attentive Context Clustering WSI), a method for Word Sense Induction, suitable for languages with limited resources. Pretrained on a small corpus and given an ambiguous word (a query word) and a set of excerpts that contain it, ACCWSI uses an attention mechanism for generating context-aware embeddings, distinguishing between the different senses assigned to the query word. These embeddings are then clustered to provide groups of main common uses of the query word. We show that ACCWSI performs well on the SemEval-2 2010 WSI task. ACCWSI also demonstrates practical applicability for shedding light on the meanings of ambiguous words in ancient languages, such as Classical Hebrew and Akkadian. In the near future, we intend to turn ACCWSI into a practical tool for linguists and historians.

Biomedical Word Sense Disambiguation Based on Graph Attention Networks

Biomedical words have many semantics. Biomedical word sense disambiguation (WSD) is an important research issue in biomedicine field. Biomedical WSD refers to the process of determining meanings of ambiguous word according to its context. It is widely applied to process, translate and retrieve biomedical texts now. In order to improve WSD accuracy in biomedicine, this paper proposes a new WSD method based on graph attention neural network (GAT). Words, parts of speech, and semantic categories in context of ambiguous word are used as disambiguation features. Disambiguation features and the sentence are used as nodes to construct WSD graph. GAT is used to extract discriminative features, and softmax function is applied to determine semantic category of biomedical ambiguous word. MSH dataset is used to optimize GAT-based WSD classifier and test its accuracy. Experiments show that average accuracy of the proposed method is improved. At the same time, majority voting strategy is adopted to optimize GAT-based WSD classifier further.

The impact of learning new meaning on the previously learned meaning of L2 ambiguous words: The role of semantic similarity

Previous study has found that previously learned meaning affects the learning of new meaning for known second language (L2) words. However, it is not clear whether learning a new meaning also affects the previously learned meaning and whether this effect is modulated by the semantic similarity between them. The current study aimed to explore this issue using event-related potential technique. A word learning task was used, in which Chinese-English bilinguals were required to learn a new meaning that was semantically related or unrelated to the previously learned meaning of familiar L2 words and judge the semantic relatedness between the trained word and a probe word in the same trial. The results showed that both the N400 and late-positive component (LPC) amplitudes for probe words in the unrelated new meaning condition were significantly different from the unlearned condition, which suggests learning new meaning interferes with accessing the previously learned meaning. Moreover, significantly less positive LPC was found in the unrelated new meaning condition relative to the related new meaning condition, showing a mediation of semantic similarity in the perturbation effect. We conclude that learning L2 new meaning has a backward interference effect on accessing the previously learned meaning and this effect is modulated by semantic similarity. These findings provide supporting evidence for the interaction mechanism of learning the multiple meanings of L2 ambiguous words.

Emotionality effects in ambiguous word recognition: The crucial role of the affective congruence between distinct meanings of ambiguous words.

There is substantial evidence that affectively charged words (e.g., party or gun) are processed differently from neutral words (e.g., pen), although there are also inconsistent findings in the field. Some lexical or semantic variables might explain such inconsistencies, due to the possible modulation of affective word processing by these variables. The aim of the present study was to examine the extent to which affective word processing is modulated by semantic ambiguity. We conducted a large lexical decision study including semantically ambiguous words (e.g., cataract) and semantically unambiguous words (e.g., terrorism), analysing the extent to which reaction times (RTs) were influenced by their affective properties. The findings revealed a valence effect in which positive valence made RTs faster, whereas negative valence slowed them. The valence effect diminished as the semantic ambiguity of words increased. This decrease did not affect all ambiguous words, but was observed mainly in ambiguous words with incongruent affective meanings. These results highlight the need to consider the affective properties of the distinct meanings of ambiguous words in research on affective word processing.

Verb Sense Disambiguation by Measuring Semantic Relatedness between Verb and Surrounding Terms of Context

Word sense disambiguation (WSD) is considered an AI complete problem which may be defined as the ability to resolve the intended meaning of ambiguous words occurring in a language. Language has complex structure and is highly ambiguous which has deep rooted relations between its different components specifically words, sentences and paragraphs. Incidentally, human beings can easily comprehend and resolve the intended meanings of the ambiguous words. The difficulty arises in building a highly accurate machine translation system or information retrieval system because of ambiguity. A number of algorithms have been devised to solve ambiguity but the success rate of these algorithms are very much limited. Context might have played a decisive role in human judgment while deciphering the meaning of polysemic words. A significant number of psychological models have been proposed to emulate the way the human beings understand the meaning of words, sentences or text depending on the context. The pertinent question that the researchers want to address is how the meanings are represented by human beings in mental memory and whether it is feasible to simulate with a computational model. Latent Semantic Analysis (LSA), a mathematical technique which is effective in representation of meanings in the form of vectors that closely approximates human semantic space. By comparing the vectors in the LSA generated semantic space, the closest neighbours of the word vector can be derived which indirectly provides lot of information about a word. However, LSA does not provide a complete theory of meaning. That is why psychological process modules are combined with LSA to make the theory of meaning concrete. Predication algorithm with LSA was proposed by Kintch, 2001 which was sufficient to capture various word senses and was successful in homonym disambiguation. Meaning of a word might have multiple senses specifically verbs. For example, verb “run” has 42 senses in WordNet. In order to find the correct sense of a verb is really a daunting task and resolving verb ambiguity using psycholinguistic model is very much limited. The proposed method has exploited the high dimensional vector LSA space resulted from training samples by applying predication algorithm to derive the most appropriate semantic neighbours for the target polysemous verb from the semantic space. Finally the vector space of test samples are checked with the training samples i.e. semantic neighbours to classify the senses of polysemous words in accurate manner.

МОЖЛИВОСТІ ВИКОРИСТАННЯ КОНТЕКСТУАЛЬНОГО МЕТОДУ ПРИ ВИВЧЕННІ ЛЕКСИКОЛОГІЇ ТУРЕЦЬКОЇ МОВИ

The relevance of the Turkish Language learning is substantiated in the article for many reasons. First, because of its prevalence in the whole world. Secondly, due to the huge number of lexical and stylistic features, such as context, polysemantic words, slang, dialects, colloquial words, phraseological units and their diversity. Thirdly, because of its morphological features and also thanks to another pleasant feature of the Turkish language, which is closely related to its clarity and conciseness. The article defines the possibilities of the contextual method using in the Turkish language Lexicology studying. Such teaching methods as descriptive (for a general description of the context); contextual-interpretive (to identify the functional and semantic meaning of a word), as well as a method of creating a problem situation using a contextual task were used for achieving the goal. The essence of the term “context” is determined and the place of context in modern linguistics is considered. The emphasis was also placed on the problems of general Lexicology, which allow the use of contextual analysis and substantiated the primary role of the context in its different types in identifying meanings. The classification of the context is considered, namely the narrow context (microcontext) and wide context (macrocontext). The difficulties of translating words from Turkish and vice versa due to their ambiguity are stipulated. Context has been shown to understand the meaning of a word or phrase. It is a powerful weapon that sheds light on the meaning of unfamiliar and ambiguous words, and the semantic characteristics of the word are described, traced, studied and analyzed. They are based on a typical context. This feature is peculiar not only any foreign language, but to Turkish due to its emotional color and individuality.

What is semantic diversity and why does it facilitate visual word recognition?

Previous research has speculated that semantic diversity and lexical ambiguity may be closely related constructs. Our research sought to test this claim in respect of the semantic diversity measure proposed by Hoffman et al. (2013). To this end, we replicated the procedure described by Hoffman et al., Behavior Research Methods, 45(3), 718–730 (2013) for computing multidimensional representations of contextual information using Latent Semantic Analysis, and from these we derived semantic diversity values for 28,555 words. We then replicated the facilitatory effect of semantic diversity on word recognition using existing data resources and observed this effect to be greater for low-frequency words. Yet, we found no relationship between this measure and lexical ambiguity effects in word recognition. Further analysis of the LSA-based contextual representations used to compute Hoffman et al. (2013) measure of semantic diversity revealed that they do not capture the distinct meanings of ambiguous words. Instead, these contextual representations appear to capture general information about the topics and types of written material in which words occur. These analyses suggest that the semantic diversity metric previously proposed by Hoffman et al. (2013) facilitates word recognition because high-diversity words are likely to have been encountered no matter what one has read, whereas many participants may not have encountered lower-diversity words simply because the topics and types of written material in which they occur are more restricted.

Semantic Processing in Autism Spectrum Disorders Is Associated With the Timing of Language Acquisition: A Magnetoencephalographic Study.

Individuals with autism show difficulties in using sentence context to identify the correct meaning of ambiguous words, such as homonyms. In this study, the brain basis of sentence context effects on word understanding during reading was examined in autism spectrum disorder (ASD) and typical development (TD) using magnetoencephalography. The correlates of a history of developmental language delay in ASD were also investigated. Event related field responses at early (150 ms after the onset of a final word) and N400 latencies are reported for three different types of sentence final words: dominant homonyms, subordinate homonyms, and unambiguous words. Clear evidence for semantic access was found at both early and conventional N400 latencies in both TD participants and individuals with ASD with no history of language delay. By contrast, modulation of evoked activity related to semantic access was weak and not significant at early latencies in individuals with ASD with a history of language delay. The reduced sensitivity to semantic context in individuals with ASD and language delay was accompanied by strong right hemisphere lateralization at early and N400 latencies; such strong activity was not observed in TD individuals and individuals with ASD without a history of language delay at either latency. These results provide new evidence and support for differential neural mechanisms underlying semantic processing in ASD, and indicate that delayed language acquisition in ASD is associated with different lateralization and processing of language.

Lexical Ambiguity Processing in Persian-speaking Patients With Broca and Wernicke Aphasia

Introduction: Wernicke and Broca are two essential types of aphasia in which patients’ productive and comprehensive abilities are affected, respectively. Although the lexico-semantic knowledge, as the heart of language comprehension, has been investigated in many research studies, there are still some controversies regarding the nature of probable lexical deficits in these patients. This study tried to delve into this issue and provide a plausible explanation in the Persian setting.Materials and Methods: In doing so, 6 patients with Broca Aphasia (BA), 6 patients with Wernicke Aphasia (WA), and 12 healthy age- and education-matched monolingual controls were selected. Conducting a lexical decision task, each patient was required to decide whether the third word of an auditory presented triplet was meaningful or not. The first and last words of the triplet were related or unrelated to the ambiguous middle word.Results: The results showed the similarity of the performance of WA patients to that of healthy control. That is, the context shaped by the first word facilitated the activation of the third word. Thus, they exhibited selected access to different meanings of ambiguous words as the healthy controls did. In contrast, semantic facilitation was not observed in BA patients.Conclusion: Our results supported the previous findings asserting the intactness of semantic representation in WA patients.

HEROÍNA: Drug or hero? Meaning-dependent valence norms for ambiguous Spanish words

AbstractSemantically ambiguous and emotional words occur frequently in language, and the different meanings of ambiguous words can sometimes have different emotional loads. For example, the Spanish word heroína (heroin/heroine) can refer to a drug or to a woman who performs a heroic act. Because both ambiguity and emotionality affect word processing, there is a need for normative databases that include data on the emotionality of the different meanings of such words. Thus far, no bases of this type are available in Spanish. With this in mind, the current study will present meaning-dependent affective (valence) ratings for 252 Spanish ambiguous words. The analyses performed show that (a) among ambiguous words, those words with meanings that have distinct affective valence are quite frequent, (b) ambiguous words rated as neutral in isolation can have meanings of opposite valence (i.e., negative-positive or positive-negative), and (c) the valence estimated for ambiguous words in isolation is better explained by the weighted average of the valence of their meanings by dominance. A database of this kind can be useful both for basic research (e.g., relationship between emotion and language and ambiguity processing) and for applied research (e.g., cognitive and emotional biases in emotional disorders and second language learning).

Emotion and location cues bias conceptual retrieval in people with deficient semantic control

Visuo-spatial context and emotional valence are powerful cues to episodic retrieval, but the contribution of these inputs to semantic cognition has not been widely investigated. We examined the impact of visuo-spatial, facial emotion and prosody cues and miscues on the retrieval of dominant and subordinate meanings of ambiguous words. Cue photographs provided relevant visuo-spatial or emotional information, consistent with the interpretation of the ambiguous word being probed, while miscues were consistent with an alternative interpretation. We compared the impact of these cues in healthy controls and semantic aphasia patients with deficient control over semantic retrieval following left-hemisphere stroke. Patients showed greater deficits in retrieving the subordinate meanings of ambiguous words, and stronger effects of cueing and miscuing relative to healthy controls. These findings suggest that contextual cues that guide retrieval to the appropriate semantic information reduce the need to constrain semantic retrieval internally, while miscues that are not aligned with the task increase the need for semantic control. Moreover, both valence and visuo-spatial context can prime particular semantic interpretations, in line with theoretical frameworks that argue meaning is computed through the integration of these features. In semantic aphasia, residual comprehension relies heavily on facial expressions and visuospatial cues. This has important implications for patients, their families and clinicians when developing new or more effective modes of communication.

Developing Corpora using Wikipedia and Word2vec for Word Sense Disambiguation

Word Sense Disambiguation (WSD) is one of the most difficult problems in the artificial intelligence field or well known as AI-hard or AI-complete. A lot of problems can be solved using word sense disambiguation approaches like sentiment analysis, machine translation, search engine relevance, coherence, anaphora resolution, and inference. In this paper, we do research to solve WSD problem with two small corpora. We propose the use of Word2vec and Wikipedia to develop the corpora. After developing the corpora, we measure the sentence similarity with the corpora using cosine similarity to determine the meaning of the ambiguous word. Lastly, to improve accuracy, we use Lesk algorithms and Wu Palmer similarity to deal with problems when there is no word from a sentence in the corpora (we call it as semantic similarity). The results of our research show an 86.94% accuracy rate and the semantic similarity improve the accuracy rate by 12.96% in determining the meaning of ambiguous words.

Listeners and readers generalize their experience with word meanings across modalities.

Research has shown that adults’ lexical-semantic representations are surprisingly malleable. For instance, the interpretation of ambiguous words (e.g., bark) is influenced by experience such that recently encountered meanings become more readily available (Rodd et al., 2016, 2013). However, the mechanism underlying this word-meaning priming effect remains unclear, and competing accounts make different predictions about the extent to which information about word meanings that is gained within one modality (e.g., speech) is transferred to the other modality (e.g., reading) to aid comprehension. In two Web-based experiments, ambiguous target words were primed with either written or spoken sentences that biased their interpretation toward a subordinate meaning, or were unprimed. About 20 min after the prime exposure, interpretation of these target words was tested by presenting them in either written or spoken form, using word association (Experiment 1, N = 78) and speeded semantic relatedness decisions (Experiment 2, N = 181). Both experiments replicated the auditory unimodal priming effect shown previously (Rodd et al., 2016, 2013) and revealed significant cross-modal priming: primed meanings were retrieved more frequently and swiftly across all primed conditions compared with the unprimed baseline. Furthermore, there were no reliable differences in priming levels between unimodal and cross-modal prime-test conditions. These results indicate that recent experience with ambiguous word meanings can bias the reader’s or listener’s later interpretation of these words in a modality-general way. We identify possible loci of this effect within the context of models of long-term priming and ambiguity resolution.

Eye Movements in English Lexical Ambiguity Resolution

English Language ambiguity has always been one of the major concerns of psycholinguistic research. A number of studies have been carried out to investigate different types of ambiguity resolution, including phonological ambiguity resolution, lexical ambiguity resolution, syntactic ambiguity resolution and pragmatic ambiguity resolution. Studies of lexical ambiguity resolution have found that lexically ambiguous words with two meanings of the same frequency took longer to read than lexically unambiguous words, suggesting that different meanings of ambiguous words compete for integration to the local context. This paper gives an overview of the classical psycholinguistic studies on English lexical ambiguity resolution which employed the eye-tracking technology.

EFFECT OF ASYMMETRY OF SEMANTIC ACTIVATION IN UNCONSCIOUS AWARENESS OF AMBIGUOUS LEXIS

The article addresses the issue of perception and processing of ambiguous information. Experimental effects of understanding of polysemantic stimuli (homonyms, reversed figures) on conscious and unconscious levels have been analyzed. It is shown that implicit awareness of several meanings occurs before the explication of one meaning of stimulus take place. The aim of the current research is to discover the effect of asymmetry of semantic activation in unconscious perception of ambiguous information (in homonyms). The procedures and results of the two conducted studies are fully described. In Experiment 1, technique of priming was used, in which ambiguous words were used as prime stimuli. Participants were randomly divided into four groups (two experimental and two control groups). Unconscious primes with visual mask were demonstrated to participants of experimental groups. Then, two words were demonstrated, only one of which had semantic relation to one meaning of the prime stimulus. The sets of stimuli in the experimental groups differed only in the words that were semantically related to the meaning of the ambiguous prime. The words that were not semantically related to the prime stimuli were identical. Then all partici pants (including participants in control groups) were told to choose as quickly as they could one word of two, using keys “left” or “right”. The results supported the hypothesis about the asymmetry of semantic activation in general. Participants did choose words related to particular meaning of prime stimuli more often.In Experiment 2, participants were asked to write down associations with the same words that were used as primes in Experiment 1. The results have shown that stronger association between prime and ambiguous stimuli facilitates priming-effect, while weaker association makes priming-effect insignificant.The present research supports the proposition that all alternative meanings of ambiguous words are more or less actualized implicitly, which influences further cognitive activity. The described phenomenon was labeled as the effect of semantic activation asymmetry. On the unconscious level, this asymmetry shows up as quantitative difference between priming-effects. On the conscious level, this asymmetry shows up as easiness and accessibility of verbal associations on corresponding meaning of the ambiguous word.