Words In Context Research Articles

A concert by Theresa “Bear” Fox

Theresa “Bear” Fox is a prolific singer and songwriter from the Mohawk Nation at Akwesasne. During this concert, she will present songs from her eight CD albums highlighting Akwesasne culture. The ancestral land of the Akwesasne community of about 12,000 people spreads across both banks of the St. Lawrence River and intersects Quebec, Ontario, and New York State. Bear Fox's music is deeply rooted in her living indigenous tradition. Following Akwesasne culture, Bear Fox's songs often have a specific function. Some songs are healing songs, and others are provided as gifts to family members. Bear Fox writes her lyrics in both Mohawk language and English, which will give the audience the opportunity to listen to the authentic sound of the Akwesasne music tradition and directly experience the meaning of words in the larger context of Native American songs. During the concert, Bear Fox will explain how each song has a story that is grown from a planted seed. Songs are often written to honor Mother Earth, mothers, water, wind, Grandmother Moon, and our newborns. Bear Fox believes in the positive message that can come from songs, which is a central part of the Mohawk culture.

Probing the Representational Structure of Regular Polysemy via Sense Analogy Questions: Insights from Contextual Word Vectors.

Regular polysemes are sets of ambiguous words that all share the same relationship between their meanings, such as CHICKEN and LOBSTER both referring to an animal or its meat. To probe how a distributional semantic model, here exemplified by bidirectional encoder representations from transformers (BERT), represents regular polysemy, we analyzed whether its embeddings support answering sense analogy questions similar to "is the mapping between CHICKEN (as an animal) and CHICKEN (as a meat) similar to that which maps between LOBSTER (as an animal) to LOBSTER (as a meat)?" We did so using the LRcos model, which combines a logistic regression classifier of different categories (e.g., animal vs. meat) with a measure of cosine similarity. We found that (a) the model was sensitive to the shared structure within a given regular relationship; (b) the shared structure varies across different regular relationships (e.g., animal/meat vs. location/organization), potentially reflective of a "regularity continuum;" (c) some high-order latent structure is shared across different regular relationships, suggestive of a similar latent structure across different types of relationships; and (d) there is a lack of evidence for the aforementioned effects being explained by meaning overlap. Lastly, we found that both components of the LRcos model made important contributions to accurate responding and that a variation of this method could yield an accuracy boost of 10% in answering sense analogy questions. These findings enrich previous theoretical work on regular polysemy with a computationally explicit theory and methods, and provide evidence for an important organizational principle for the mental lexicon and the broader conceptual knowledgesystem.

From unstructured accident reports to a hybrid decision support system for occupational risk management: The consensus converging approach

Introduction: Workplace accidents in the petroleum industry can cause catastrophic damage to people, property, and the environment. Earlier studies in this domain indicate that the majority of the accident report information is available in unstructured text format. Conventional techniques for the analysis of accident data are time-consuming and heavily dependent on experts’ subject knowledge, experience, and judgment. There is a need to develop a machine learning-based decision support system to analyze the vast amounts of unstructured text data that are frequently overlooked due to a lack of appropriate methodology. Method: To address this gap in the literature, we propose a hybrid methodology that uses improved text-mining techniques combined with an un-bias group decision-making framework to combine the output of objective weights (based on text mining) and subjective weights (based on expert opinion) of risk factors to prioritize them. Based on the contextual word embedding models and term frequencies, we extracted five important clusters of risk factors comprising more than 32 risk sub-factors. A heterogeneous group of experts and employees in the petroleum industry were contacted to obtain their opinions on the extracted risk factors, and the best-worst method was used to convert their opinions to weights. Conclusions and Practical Applications: The applicability of our proposed framework was tested on the data compiled from the accident data released by the petroleum industries in India. Our framework can be extended to accident data from any industry, to reduce analysis time and improve the accuracy in classifying and prioritizing risk factors.

On the Role of Morphological Information for Contextual Lemmatization

Abstract Lemmatization is a natural language processing (NLP) task that consists of producing, from a given inflected word, its canonical form or lemma. Lemmatization is one of the basic tasks that facilitate downstream NLP applications, and is of particular importance for high-inflected languages. Given that the process to obtain a lemma from an inflected word can be explained by looking at its morphosyntactic category, including fine-grained morphosyntactic information to train contextual lemmatizers has become common practice, without considering whether that is the optimum in terms of downstream performance. In order to address this issue, in this article we empirically investigate the role of morphological information to develop contextual lemmatizers in six languages within a varied spectrum of morphological complexity: Basque, Turkish, Russian, Czech, Spanish, and English. Furthermore, and unlike the vast majority of previous work, we also evaluate lemmatizers in out-of-domain settings, which constitutes, after all, their most common application use. The results of our study are rather surprising. It turns out that providing lemmatizers with fine-grained morphological features during training is not that beneficial, not even for agglutinative languages. In fact, modern contextual word representations seem to implicitly encode enough morphological information to obtain competitive contextual lemmatizers without seeing any explicit morphological signal. Moreover, our experiments suggest that the best lemmatizers out-of-domain are those using simple UPOS tags or those trained without morphology and, lastly, that current evaluation practices for lemmatization are not adequate to clearly discriminate between models.

AN ANALYSIS OF ASSOCIATIVE MEANING FROM “ALADDIN” (2019) SOUNDTRACKS LYRICS

This paper reviews An Analysis of Associative Meaning From "Aladdin" (2019) Soundtracks Lyrics. This research aims to find meaning in each Aladdin soundtrack as well as the meaning of the lyrics using associative meaning theory. This kind of research is qualitative. The source of this research data is the result of the search for meaning in each existing song, then the result is loaded in the form of tables or data. The result that has been obtained; it can be concluded that the data has associative meaning based on Leech's theory of associative meaning. The results showed that there are 89 data and 4 types of associative meanings found in the 6 soundtracks in the Aladdin movie, 4 of them are; Connotative meaning contains 25 data, Social or Stylistic Meaning contains 24 data, Affective meaning has 23 data, while Collective and Reflective Meaning has 17 data. This study used a qualitative descriptive method in analyzing the data. Leech's theory is used in this study to analyze associative meaning. Implications on Leech's theory are, by observing the collocation of words in lyrics and analysing how these combinations of words form specific associations in the listener's mind. Leech emphasizes the importance of relationships between words in a particular context and how such relationships can form broader meaning, and also by studying lyrics and analysing them by using the Leech Theory framework, readers can develop skills in understanding and analyzing English contextually, not just literally.

Dilated convolution for enhanced extractive summarization: A GAN-based approach with BERT word embedding

Text summarization (TS) plays a crucial role in natural language processing (NLP) by automatically condensing and capturing key information from text documents. Its significance extends to diverse fields, including engineering, healthcare, and others, where it offers substantial time and resource savings. However, manual summarization is a laborious task, prompting the need for automated text summarization systems. In this paper, we propose a novel strategy for extractive summarization that leverages a generative adversarial network (GAN)-based method and Bidirectional Encoder Representations from Transformers (BERT) word embedding. BERT, a transformer-based architecture, processes sentence bidirectionally, considering both preceding and following words. This contextual understanding empowers BERT to generate word representations that carry a deeper meaning and accurately reflect their usage within specific contexts. Our method adopts a generator and discriminator within the GAN framework. The generator assesses the likelihood of each sentence in the summary while the discriminator evaluates the generated summary. To extract meaningful features in parallel, we introduce three dilated convolution layers in the generator and discriminator. Dilated convolution allows for capturing a larger context and incorporating long-range dependencies. By introducing gaps between filter weights, dilated convolution expands the receptive field, enabling the model to consider a broader context of words. To encourage the generator to explore diverse sentence combinations that lead to high-quality summaries, we introduce various noises to each document within our proposed GAN. This approach allows the generator to learn from a range of sentence permutations and select the most suitable ones. We evaluate the performance of our proposed model using the CNN/Daily Mail dataset. The results, measured using the ROUGE metric, demonstrate the superiority of our approach compared to other tested methods. This confirms the effectiveness of our GAN-based strategy, which integrates dilated convolution layers, BERT word embedding, and a generator-discriminator framework in achieving enhanced extractive summarization performance.

A Semantics-enhanced Topic Modelling Technique: Semantic-LDA

Topic modelling is a beneficial technique used to discover latent topics in text collections. But to correctly understand the text content and generate a meaningful topic list, semantics are important. By ignoring semantics, that is, not attempting to grasp the meaning of the words, most of the existing topic modelling approaches can generate some meaningless topic words. Even existing semantic-based approaches usually interpret the meanings of words without considering the context and related words. In this article, we introduce a semantic-based topic model called semantic-LDA that captures the semantics of words in a text collection using concepts from an external ontology. A new method is introduced to identify and quantify the concept–word relationships based on matching words from the input text collection with concepts from an ontology without using pre-calculated values from the ontology that quantify the relationships between the words and concepts. These pre-calculated values may not reflect the actual relationships between words and concepts for the input collection, because they are derived from datasets used to build the ontology rather than from the input collection itself. Instead, quantifying the relationship based on the word distribution in the input collection is more realistic and beneficial in the semantic capture process. Furthermore, an ambiguity handling mechanism is introduced to interpret the unmatched words, that is, words for which there are no matching concepts in the ontology. Thus, this article makes a significant contribution by introducing a semantic-based topic model that calculates the word–concept relationships directly from the input text collection. The proposed semantic-based topic model and an enhanced version with the disambiguation mechanism were evaluated against a set of state-of-the-art systems, and our approaches outperformed the baseline systems in both topic quality and information filtering evaluations.

Assessing the Impact of Static, Contextual and Character Embeddings for Arabic Machine Translation

Word embeddings/representations are an important component of Natural Language Processing (NLP) tasks. Most Neural Machine Translation (NMT) systems that use such representations disregard word morphology by assigning a unique vector to each unique word in the used vocabulary and thus cannot handle Out-Of-Vocabulary (OOV) words. In some languages, such as Arabic, the meaning of words is associated with the meaning of the individual characters that constitute them, as these characters embody internal information. In this study, a combination of character- and word- level models is used to determine the most effective approaches to semantically and morphologically representing affective Arabic words. Furthermore, this work examines the strategy of combining static, character and contextual word embeddings to obtain richer representations for the Arabic Machine Translation (MT) task. To the best of our knowledge, we are the first to investigate the combination of static word embeddings, contextual ones and character-level representation in Arabic MT. Furthermore, a Deep Learning (DL) architecture is employed on data preprocessed by various prominent preprocessing techniques. Various experiments were conducted and the findings indicate that the integration of various models for word embedding and character-level representation is feasible and more effective than the state-of-the-art Arabic MT systems.

ID-SF-Fusion: a cooperative model of intent detection and slot filling for natural language understanding

PurposeIntent detection (ID) and slot filling (SF) are two important tasks in natural language understanding. ID is to identify the main intent of a paragraph of text. The goal of SF is to extract the information that is important to the intent from the input sentence. However, most of the existing methods use sentence-level intention recognition, which has the risk of error propagation, and the relationship between intention recognition and SF is not explicitly modeled. Aiming at this problem, this paper proposes a collaborative model of ID and SF for intelligent spoken language understanding called ID-SF-Fusion.Design/methodology/approachID-SF-Fusion uses Bidirectional Encoder Representation from Transformers (BERT) and Bidirectional Long Short-Term Memory (BiLSTM) to extract effective word embedding and context vectors containing the whole sentence information respectively. Fusion layer is used to provide intent–slot fusion information for SF task. In this way, the relationship between ID and SF task is fully explicitly modeled. This layer takes the result of ID and slot context vectors as input to obtain the fusion information which contains both ID result and slot information. Meanwhile, to further reduce error propagation, we use word-level ID for the ID-SF-Fusion model. Finally, two tasks of ID and SF are realized by joint optimization training.FindingsWe conducted experiments on two public datasets, Airline Travel Information Systems (ATIS) and Snips. The results show that the Intent ACC score and Slot F1 score of ID-SF-Fusion on ATIS and Snips are 98.0 per cent and 95.8 per cent, respectively, and the two indicators on Snips dataset are 98.6 per cent and 96.7 per cent, respectively. These models are superior to slot-gated, SF-ID NetWork, stack-Prop and other models. In addition, ablation experiments were performed to further analyze and discuss the proposed model.Originality/valueThis paper uses word-level intent recognition and introduces intent information into the SF process, which is a significant improvement on both data sets.

Using music reviews in the intermediate L2 German classroom: An exploratory lesson in genre‐based writing

AbstractThis article presents an exploratory genre‐based writing lesson conducted in a third‐year collegiate German language course. While genre‐based approaches to writing have gained recognition in second language (L2) contexts, little attention has been given to learner perceptions of genre‐based writing activities. Additionally, there is a lack of pedagogical strategies and materials for intermediate‐level German language classes. This article addresses these gaps by introducing a lesson on consumer music reviews designed following genre‐based pedagogical principles, the flipped classroom approach, and the use of language corpora. The lesson, consisting of two 75‐min class periods, aimed at familiarizing students with the structure and style of consumer music reviews, culminating in the writing of their own reviews. Using authentic texts and incorporating the Digitales Wörterbuch der deutschen Sprache corpus tool for vocabulary building, students engaged with contemporary German‐language music. Data from student surveys indicated positive perceptions of the lesson, particularly regarding vocabulary acquisition, lesson structure, and topic relevance. Students also found the corpus tool beneficial for contextual understanding and word usage. This exploratory lesson highlights the potential of genre‐based writing pedagogy, the flipped classroom model, and corpus tools for L2 writing development and contributes to the growing body of literature on genre‐based writing in L2 classrooms.

Childhood conduct problems and parent–child talk during social and nonsocial play contexts: a naturalistic home-based experiment

Parent–child interactions are a critical pathway to emotion socialization, with disruption to these processes associated with risk for childhood behavior problems. Using computational linguistics methods, we tested whether (1) play context influenced parent–child socioemotional language, and (2) child conduct problems or callous-unemotional traits were associated with patterns of socioemotional or nonsocial language across contexts. Seventy-nine parent–child dyads (children, 5–6 years old) played a socioemotional skills (“social context”) or math (“nonsocial context”) game at home. We transcribed and analyzed game play, which had been audio recorded by participants. The social context elicited more socioemotional and cognitive words, while the nonsocial context elicited more mathematical words. The use of socioemotional language by parents and children was more strongly correlated in the social context, but context did not moderate the degree of correlation in cognitive or mathematical word use between parents and children. Children with more conduct problems used fewer socioemotional words in the social context, while children with higher callous-unemotional traits used fewer cognitive words in both contexts. We highlight the role of context in supporting socioemotionally rich parent–child language interactions and provide preliminary evidence for the existence of linguistic markers of child behavior problems. Our results also inform naturalistic assessments of parent–child interactions and home-based interventions for parents and children facing socioemotional or behavioral challenges.

Effective history and efficient integration for conversational machine reading comprehension

The task of conversational machine reading comprehension (CMRC) is an extension of single-turn machine reading comprehension to multi-turn settings, to relflect the conversational way in which people seek information. The correlations between multiple rounds of questions mean that the conversation history is critical to solving the CMRC task. However, existing CMRC models ignore the interference that arises from using excessive historical information to answer the current question when incorporating the dialogue history into the current question. In this paper, an effective Question Selection Module (QSM) is designed to select most relevant historical dialogues when answering the current question through question coupling and coarse-to-fine matching. In addition, most existing approaches perform memory inference by stacked RNNs at context word level, without considering semantic information flowing in the direction of conversation flow. In view of this problem, we implement sequential recurrent reasoning at the turn level of the dialogue, where the turn information contains all the filtered historical semantics for the current step. We conduct experiments on two benchmark datasets, QuAC and CoQA, released by Stanford University. The results confirm that our model satisfactorily captures the valid history and performs recurrent reasoning, and our model achieves an F1-score of 83.0% on CoQA dataset and 67.8% on QuAC dataset, outperforming the best alternative model by 4.6% on CoQA and 2.7% on QuAC.

Text clustering based on pre-trained models and autoencoders.

Text clustering is the task of grouping text data based on similarity, and it holds particular importance in the medical field. sIn healthcare, medical data clustering is a highly active and effective research area. It not only provides strong support for making correct medical decisions from medical datasets but also aids in patient record management and medical information retrieval. With the development of the healthcare industry, a large amount of medical data is being generated, and traditional medical data clustering faces significant challenges. Many existing text clustering algorithms are primarily based on the bag-of-words model, which has issues such as high dimensionality, sparsity, and the neglect of word positions and context. Pre-trained models are a deep learning-based approach that treats text as a sequence to accurately capture word positions and context information. Moreover, compared to traditional K-means and fuzzy C-means clustering models, deep learning-based clustering algorithms are better at handling high-dimensional, complex, and nonlinear data. In particular, clustering algorithms based on autoencoders can learn data representations and clustering information, effectively reducing noise interference and errors during the clustering process. This paper combines pre-trained language models with deep embedding clustering models. Experimental results demonstrate that our model performs exceptionally well on four public datasets, outperforming most existing text clustering algorithms, and can be applied to medical data clustering.

Sentence-level sentiment classification based on multi-attention bidirectional gated spiking neural P systems

The GSNP model is a new recurrent-like network inspired by nonlinear spiking mechanisms in nonlinear spiking neural P systems. In this study, a novel sentiment classification model MA-BiGSNP is established by using bidirectional GSNP model combined with multi-attention mechanism. BiGSNP, which is created by two GSNP models with opposite directions, is used to capture semantic correlations between word contexts in a sentence. The multi-attention mechanism simulates the variety of relationships between sentences as well as the significance of words in sentences. To evaluate the effectiveness of the proposed MA-BiGSNP model, we perform comparative experiments and ablation experiments on five real datasets and twelve baseline models. The experimental results show that the proposed MA-BiGSNP model is effective for sentiment classification task.

Contextual Word Embedding for Biomedical Knowledge Extraction: a Rapid Review and Case Study.

The online version contains supplementary material available at 10.1007/s41666-023-00157-y.

Students Responses Toward the Use of Wordwall Unjumbled Word in Teaching Learning Context

English is the most influential language in the world. However, since learning a language is rarely applied in real life, students tend to think that learning this language is unnecessary. In this situation, integrating technology and education education becomes the most crucial part of making the learning process more successful. This article aims to determine the students' responses to using the educational online game Wordwall Unjumbled Word. Qualitative descriptive is used in this research’s design. As the object of the research, the researcher chose 35 students from 11th graders of Social 4 in 10 senior high schools in Semarang. The results of the research are 57,1% agreed that they like the media, 33,6% agreed that the media is enjoyable, 51,4% agreed the media improves their motivation, 57,1% agreed that the media draws their interest, 48,6% agreed that the media made their learning easier, 34,3% stated that the media did not make them stressed, 42,9% agreed that understand the material, 46,7% agreed that the media is easy to be used, 40% agreed that they would use this media in the subsequent learning and 45,7% agreed that the media is appropriate for English learning. The results found in this research should be considered when choosing a suitable media for the English teaching and learning process.

Syntactic Graph Attention Network for Aspect-Level Sentiment Analysis

Aspect-level sentiment classification (ASC) is designed to identify the sentiment orientation of given aspect terms in a sentence. Previous neural networks have used attention mechanisms to align context words with the appropriate aspect terms. Without considering syntactic dependencies, these models may erroneously focus on context words that are not related to the aspect terms. To address this issue, the graph convolution network (GCN) and the graph attention network (GAT) are proposed to build a graph based on the dependency parse tree, allowing the representations of context words to be propagated to the aspect terms according to their syntactic dependencies. However, these models consider all syntactic dependencies to be of the same type, and thus may result in inappropriate propagation of word representations in the graph. To further distinguish between the syntactic dependencies, this study proposes a syntactic graph attention network (SGAN) to incorporate the knowledge of dependency types into the graph attention network. The dependency types are modeled as edge embeddings to learn the attention weight of each edge according to its dependency type. By considering different dependency types and their weights, the proposed method can block inappropriate propagation to better associate the context words to aspect terms. To increase training process stability and enrich the diversity of graph representations, a weighted multihead attention is applied to compose the graph representations generated by different heads. The experimental results on five benchmark datasets show that the SGAN yields more accurate results than existing methods.

Sachant que : un marqueur (plus ou moins juste) de mémoire sémanticodiscursive

A construction with epistemic value, sachant que is a discourse marker that covers both semantic-logical and discursive-argumentative values. Our paper sets out to follow the behaviour of this marker through a heterogeneous corpus, from the point of view of textual typology (literary texts, everyday discourse) but also in specialised language, and more specifically legal language (written and oral). The interest in analysing this marker lies in the fact that its uses, which are very frequent in contemporary French, tend to diluteitsinitialepistemic value,which refers to a known content shared by the enunciative instances. Our analysis follows the mechanism of weakening or maintaining thesemantic and argumentativeforce of this discursive unit through various types of discourse, in order to identify the correct use of the word in approprithrough various types of discourse, in order to identify the correct use of the word in appropriate contexts and with the appropriate meaning.

Rethinking Resilience: Definition, Context, and Measure

A sizeable literature treats “resilience” without defining it. Seeking to promote resilience from ubiquitous buzzword to managerially useful concept, we focus on situations in which human decision is paramount—organizational, community, and personal resilience—and exclude discussion of, e.g., resilience of fixed infrastructure and of unmanaged natural ecosystems. We argue that the word resilience, without modification or context, carries no meaning, but that strategic planning is a context that gives specific and useful meaning to resilience. Bringing in ideas from system theory, we discuss the role of anticipation, showing that resilience is not reactive; and the role of learning, showing that resilience is not a return to the status quo ante. These results provide diagnostic guidance for organizational management and for industry commentators, and a rigorous basis for further advances in resilience theory.

Development of an error correction algorithm for Kazakh language

This article discusses a method for correcting spelling errors in the Kazakh language using the advantages of morphological analysis and a model based on noisy channels. To achieve this goal, modern problems of automatic processing of Kazakh textual information were analyzed, existing linguistic resources and processing systems of the Kazakh language were systematized, the basic requirements for the development of a system for analyzing Kazakh textual information based on machine learning were determined, and models and algorithms for extracting facts from unstructured and poorly structured text arrays were developed. The search function, an enhanced spelling correction algorithm, was utilized in this work and has the ability to recommend the proper spelling of the input text. The maximum editing distance, whether to include the original word when near matches are not found, and how to handle case sensitivity and exclusion based on regular expressions are all easily adjustable features of this functionality. Because of their adaptability, algorithms can be applied to a wide range of problems, from straightforward spell checks in user interfaces to intricate natural language processing assignments. Because of the way it’s designed, the search function finds possible corrections and verifies the context of words while accounting for user preferences like verbosity and ignore markers. Most modern multilingual natural language processing programs use only the graphical stage of text processing. On the other hand, semantic text analysis or analysis of the meaning of natural language is still an important problem in the theory of artificial intelligence and computational linguistics. But in order to process the grammar and semantics of multilingual information, precreated semantic and grammatical corpora of each natural language are necessary. To solve this problem, several tasks were considered and solved. These tasks included the analysis of research in the field of machine learning methods used in the processing of textual information, the existing problems of formalization and modeling of the Kazakh language, as well as the development and implementation of models, methods and algorithms for morphological and semantic analysis of texts of the Kazakh language.