Natural Language Processing

Semantic Annotation of Deverbal Nominalizations in the Spanish AnCora corpus Aina Peris CLiC-UB Gran Via,585 08007 Barcelona aina.peris@ub.edu Mariona Taulé CLiC-UB Gran Via,585 08007 Barcelona mtaule@ub.edu Horaci Rodríguez TALP -UPC Jordi Girona ...

Resumen: En este artículo se describe un nuevo recurso: AnCora-Nom, un léxico de nominalizaciones deverbales del español. Actualmente, contiene 1.655 entradas léxicas y 3.094 sentidos, donde cada sentido tiene asociado el tipo denotativo y la estructura argumental con los papeles temáticos correspondientes. Este léxico se ha extraído automáticamente a partir de la información anotada en el corpus AnCora-Es. AnCora-Nom se derivó teniendo en cuenta no sólo la información estrictamente relacionada con las nominalizaciones deverbales sino también con información morfológica y sintáctico-semántica previamente anotada en el corpus.

Despite the high accuracy offered by state-of-the-art deep natural-language models (e.g. LSTM, BERT), their application in reallife settings is still widely limited, as they behave like a black-box to the end-user. Hence, explainability is rapidly becoming a fundamental requirement of future-generation data-driven systems based on deep-learning approaches. Several attempts to fulfill the existing gap between accuracy and interpretability have been done. However, robust and specialized xAI (Explainable Artificial Intelligence) solutions tailored to deep natural-language models are still missing. We propose a new framework, named T-EBAnO, which provides innovative prediction-local and class-based model-global explanation strategies tailored to black-box deep natural-language models. Given a deep NLP model and the textual input data, T-EBAnO provides an objective, human-readable, domain-specific assessment of the reasons behind the automatic decision-making process. Specifically, the framework extracts sets of interpretable features mining the inner knowledge of the model. Then, it quantifies the influence of each feature during the prediction process by exploiting the novel normalized Perturbation Influence Relation index at the local level and the novel Global Absolute Influence and Global Relative Influence indexes at the global level. The effectiveness and the quality of the local and global explanations obtained with T-EBAnO are proved on (i) a sentiment analysis task performed by a fine-tuned BERT model, and (ii) a toxic comment classification task performed by an LSTM model.

A case based reasoning approach is introduced as a learning technique in the domain of machine translation of natural language. In our approach syntactical and semantic features are part of the cases in the case-base. To implement this, dependency analysers of sentences in the source and target languages are used. The case-base is filled with a learning mechanism that uses a parallel corpus of sentences with their translations. This case-base is used to make new translations.

In the past decade, we have observed a growing interest in using technologies such as artificial intelligence (AI), machine learning, and chatbots to provide assistance to language learners, especially in second language learning. By using AI and natural language processing (NLP) and chatbots, we can create an intelligent self-learning environment that goes beyond multiple-choice questions and/or fill in the blank exercises. In addition, NLP allows for learning to be adaptive in that it offers more than an indication that an error has occurred. It also provides a description of the error, uses linguistic analysis to isolate the source of the error, and then suggests additional drills to achieve optimal individualized learning outcomes. In this paper, we present our approach for developing an Artificial Intelligence-based Arabic Language and Speech Tutor (AI-ALST) for teaching the Moroccan Arabic dialect. The AI-ALST system is an intelligent tutor that provides analysis and assessment of students learning the Moroccan dialect at University of Arizona (UA). The AI-ALST provides a selflearned environment to practice each lesson for pronunciation training. In this paper, we present our initial experimental evaluation of the AI-ALST that is based on MFCC (Mel frequency cepstrum coefficient) feature extraction, bidirectional LSTM (Long Short-Term Memory), attention mechanism, and a cost-based strategy for dealing with class-imbalance learning. We evaluated our tutor on the word pronunciation of lesson 1 of the Moroccan Arabic dialect class. The experimental results show that the AI-ALST can effectively and successfully detect pronunciation errors and evaluate its performance by using 𝑭 𝟏 -𝐬𝐜𝐨𝐫𝐞, accuracy, precision, and recall.

This research work sets out to explore the major distinctions between Arabic and Bangla—the languages with unidentical origins. Comparing and analyzing the various features of these two languages requires huge linguistic expertise in the respective fields as it is a most complicated job for anyone to accomplish. Arabic and Bangla are two of the leading languages of the world, specially in terms of their number of speakers and the growing demands. As Arabic and Bangla are from unalike families of languages, they differ a lot in the word class, grammar, pronunciation, usage and so forth. The sentence in Arabic is divided into two types: verbal i.e. V+S+C and nominal i.e. S+C; while the typical Bangla sentence pattern is inflexion-based i.e. S+C+V. Like any other vocalized languages, Bangla has eleven vowels, but Arabic has no such vowels since Arabic alphabet is considered an ‘abjad’ (i.e. ‘أبجدية’ /aːbʤadiah/) meaning a ‘consonantal alphabet’ and so the syllable is often formed witho...

Proverbs are commonly metaphoric in nature and the mapping across domains is commonly established in proverbs. The abundance of proverbs in terms of metaphors makes them an extremely valuable linguistic resource since they can be utilized as a gold standard for various metaphor related linguistic tasks such as metaphor identification or interpretation. Besides, a collection of proverbs fromvarious languages annotated with metaphors would also be essential for social scientists to explore the cultural differences betweenthose languages. In this paper, we introduce PROMETHEUS, a dataset consisting of English proverbs and their equivalents in Italian.In addition to the word-level metaphor annotations for each proverb, PROMETHEUS contains other types of information such as the metaphoricity degree of the overall proverb, its meaning, the century that it was first recorded in and a pair of subjective questions responded by the annotators. To the best of our knowledge, this is the first m...

In this paper, we combine existing NLP techniques with minimal supervision to build memory tips according to the keyword method, a well established mnemonic device for second language learning. We present what we believe to be the first extrinsic evaluation of a creative sentence generator on a vocabulary learning task. The results demonstrate that NLP techniques can effectively support the development of resources for second language learning.

We present BRAINSUP, an extensible framework for the generation of creative sentences in which users are able to force several words to appear in the sentences and to control the generation process across several semantic dimensions, namely emotions, colors, domain relatedness and phonetic properties. We evaluate its performance on a creative sentence generation task, showing its capability of generating well-formed, catchy and effective sentences that have all the good qualities of slogans produced by human copywriters.

This paper describes the system developed for SemEval 2017 task 6: #HashTagWars -Learning a Sense of Humor. Learning to recognize sense of humor is the important task for language understanding applications. Different set of features based on frequency of words, structure of tweets and semantics are used in this system to identify the presence of humor in tweets. Supervised machine learning approaches, Multilayer perceptron and Naïve Bayes are used to classify the tweets in to three levels of sense of humor. For given Hashtag, the system finds the funniest tweet and predicts the amount of funniness of all the other tweets. In official submitted runs, we have achieved 0.506 accuracy using multilayer perceptron in subtask-A and 0.938 distance in subtask-B. Using Naïve bayes in subtask-B, the system achieved 0.949 distance. Apart from official runs, this system have scored 0.751 accuracy in subtask-A using SVM.

Automatic short answer grading (ASAG) has become part of natural language processing problems. Modern ASAG systems start with natural language preprocessing and end with grading. Researchers started experimenting with machine learning in the preprocessing stage and deep learning techniques in automatic grading for English. However, little research is available on automatic grading for Arabic. Datasets are important to ASAG, and limited datasets are available in Arabic. In this research, we have collected a set of questions, answers, and associated grades in Arabic. We have made this dataset publicly available. We have extended to Arabic the solutions used for English ASAG. We have tested how automatic grading works on answers in Arabic provided by schoolchildren in 6th grade in the context of serious games. We found out those schoolchildren providing answers that are 5.6 words long on average. On such answers, deep learning-based grading has achieved high accuracy even with limited training data. We have tested three different recurrent neural networks for grading. With a transformer, we have achieved an accuracy of 95.67%. ASAG for school children will help detect children with learning problems early. When detected early, teachers can solve learning problems easily. This is the main purpose of this research.

The term Morphologically Rich Languages (MRLs) refers to languages in which significant information concerning syntactic units and relations is expressed at word-level. There is ample evidence that the application of readily available statistical parsing models to such languages is susceptible to serious performance degradation. The first workshop on statistical parsing of MRLs hosts a variety of contributions which show that despite languagespecific idiosyncrasies, the problems associated with parsing MRLs cut across languages and parsing frameworks. In this paper we review the current state-of-affairs with respect to parsing MRLs and point out central challenges. We synthesize the contributions of researchers working on parsing Arabic, Basque, French, German, Hebrew, Hindi and Korean to point out shared solutions across languages. The overarching analysis suggests itself as a source of directions for future investigations.

Translation is the transfer of the meaning of a text from one language to another. It is a means of sharing information across languages and therefore essential for addressing information inequalities. The work of translation was originally carried out by human translators and its limitations led to the development of machine translators. Machine Translation is a subfield of computational linguistics that investigates the use of computer software to translate text or speech from one natural language to another. There are different approaches to machine translation. This paper reviews the two major approaches (single vs. hybrid) to machine translation and provides critique of existing machine translation systems with their merits and demerits. Several application areas of machine translation and various methods used in evaluating them were also discussed. Our conclusion from the reviewed literatures is that a single approach to machine translation fails to achieve satisfactory perfor...

Dialectometry is a discipline devoted to studying the variations of a language around a geographical region. One of their goals is the creation of linguistic atlases capturing the similarities and differences of the language under study around the area in question. For instance, Spanish is one of the most spoken languages across the world, but not necessarily Spanish is written and spoken in the same way in different countries. This manuscript presents a broad analysis describing lexical and semantic relationships among 26 Spanish-speaking countries around the globe. For this study, we analyze four-year of the Twitter geotagged public stream to provide an extensive survey of the Spanish language vocabularies of different countries, its distributions, semantic usage of terms, and emojis. We also offer open regional word-embedding resources for Spanish Twitter to help other researchers and practitioners take advantage of regionalized models.

This study employed a text mining method to investigate the lexical features and their dynamic changes of PhD theses across the natural sciences, social sciences and humanities. Four quantitative indices, i.e. TTR, h-point, R1 and writer's view, were employed to analyze 150 PhD theses (50 theses from each discipline). Although h-point and writer's view were found counter-intuitively to show insignificant variation across disciplines, the results of TTR and R1 did reveal sharp contrasts between theses in humanities and natural sciences. While the second half of humanities theses showed a significantly higher level of lexical diversity, indicated by higher TTR, theses in natural sciences tended to be richer in content words in the first half, indicated by a higher R1. Meanwhile, theses in social sciences seemed to be more moderate, with features lying in the middle position. This study has implications not only for the widening of applications of quantitative linguistic methods but also for academic writing (especially PhD thesis writing) instruction and practice.

Abstract
This article explores the dynamic relationship between child psychology and linguistics, focusing on how language development both reflects and influences cognitive, emotional, and social growth in children. Drawing on foundational theories from Piaget, Vygotsky, and Chomsky, it examines how language acquisition aligns with cognitive milestones and social interaction. The discussion extends to the impact of environmental factors on linguistic input, the cognitive benefits of bilingualism, and the implications of language delays and disorders. By integrating perspectives from both disciplines, the article underscores the importance of a holistic, multidisciplinary approach to supporting children’s language development and overall well-being.

na publicação dos anais do evento. A RESP é uma publicação científica multidisciplinar, com publicação em fluxo contínuo mensal, cujo objetivo é auxiliar na produção intelectual do conhecimento sobre a saúde populacional, sempre desempenhando seu papel social, político e econômicos da sociedade.

This paper presents a strategy for syntactic analysis based on the combination of two different parsing techniques: lexical syntactic tagging and phrase structure syntactic parsing. The basic proposal is to take advantage of the good results on lexical syntactic tagging to improve the whole performance of unification-based parsing. The syntactic functions attached to every word by the lexical syntactic tagging are used as head features in the unification-based grammar, and are the base for grammar rules.

One of the striking commonalities between languages is the way word frequencies are distributed. Across languages, word frequencies follow a Zipfian distribution, showing a power law relation between a word's frequency and its rank . Intuitively, this means that languages have relatively few high-frequency words and many low-frequency ones. While studied extensively, little work has explored the learnability consequences of the greater predictability of words in such distributions. Here, we propose such distributions confer a learnability advantage for word segmentation, a foundational aspect of language acquisition. We capture the greater predictability of words using the information-theoretic notion of efficiency, which tells us how predictable a distribution is relative to a uniform one. We first use corpus analyses to show that child-directed speech is similarly predictable across fifteen different languages. We then experimentally investigate the impact of distribution predictability on children and adults. We show that word segmentation is uniquely facilitated at the predictability levels found in language, compared both with uniform distributions and with skewed distributions that are less predictable than those of natural language. We further show that distribution predictability impacts learning more than distribution shape, and that learning is not improved further in distributions more predictable than natural language. These novel findings illustrate learners' sensitivity to the overall predictability of the linguistic environment; suggest that the predictability levels found in language provide an optimal environment for learning; and point to the possible role of cognitive pressures in the emergence and propensity of such distributions in language. While the world's languages differ in many respects, they share certain commonalities: these can provide crucial insight on our shared cognition and how it impacts language structure. We offer a novel learnability perspective on one of the striking commonalities between languages: the Zipfian distribution of word frequencies. We show that languages have similarly predictable distributions, and that these predictability levels are uniquely facilitative for word segmentation in children and adults. We explain their lower and upper bounds as a trade-off between the competing pressures of learnability and expressivity. These findings have far-reaching empirical and theoretical implications, illustrating learners' sensitivity to distribution predictability and pointing to role of cognitive biases in the recurrence of such skewed distributions in language. * The formula for Zipfian distributions, as extended by Mandelbrot (40), appears in Equation . It shows the relation between word's frequency -f(r) and its rank -r, with two constants that determine the shape of the distribution: α sets the steepness of the curve, and β introduces a skew which enables a better fit to natural language (1, 2).

In the current context of rapid and constant evolution of global communication and specialised discourses, the need for devising methods for ensuring both high quality levels of specialised translation and successful translation training is becoming a true challenge. Steady renewal in knowledge paradigms leads to an increase in term coinage, modifications in lexical and phraseological patterns, and accommodations in discourse conventions. This situation requires teachers in specialised translation to train future translators to develop the skills meant to help them adapt rapidly to change. The tools brought by corpus linguistics offer access to the language-in-the-making and continuously emerging knowledge fields. However, methods for their efficient exploitation in translation classes can still be improved. In the current study, we present the translation-teaching framework devised specifically for such contexts. It is based on corpus linguistics, terminology management, collaborat...

Sentiment Classification is a special task of Sentiments Analysis in which a text document is assigned into some category like positive, negative, and neutral on the basis of some subjective information contained in documents. This subjective information called as sentiment features are highly responsible for efficient sentiment classification. Thus, Feature extraction is essentially an important task for sentiment classification at any level. This study explores most relevant and crucial features for sentiment classification and groups them into seven categories, named as, Basic features, Seed word features, TF-IDF, Punctuation based features, Sentence based features, N-grams, and POS lexicons. This paper proposes two new sentence based features which are helpful in assigning the overall sentiment of contrastive sentences and on the basis of proposed features; two algorithms are developed to find the sentiment of contrastive sentences. The dataset of TripAdvisor is used to evaluate our proposed features. Obtained results are compared with several state-of-theart studies using various features on the same dataset and achieve superior performance.

Machine Translation (MT) is a programmed conversion in which computer software is utilized to convert manuscripts from one Natural Language (like English) to a different Language (such as Hindi). To process any such conversion, through human or through automatic means, the conversion must be established such that it reinstate the complete sense of a manuscript from its base (source) linguistic into the target language. In this paper, the study of prevailing evaluation systems along with assessing their performance is achieved through the similarity metrics. Moreover, the authors have also presented an improved technique of translation employing features of Natural Language Processing and consequently, to acquire an enhanced and more accurate assessing Machine Translation system, a corpus is selected and the outcomes are compared with the prevailing methods. Besides this, two well-known systems such as Google and Bing decoders are selected to inquire and to assess the study of metrics called similarity metrics through Assessment of Text Essential Characteristics score. This is found to provide more accuracy than prevailing methods. Furthermore, evaluations are tested under various metrics systems like Jaccard similarity metrics, cosine similarity metrics, and sine metrics to deliver enhanced accuracy than prevailing methods.

Asian and low-resource language information processing refers to the field of computational linguistics that aims to develop natural language processing (NLP) technologies for languages that have fewer available language resources or are less commonly spoken. This is an important field of study because many languages in Asia and other parts of the world are underrepresented in the field of NLP, which may limit access to information and technology for speakers of these languages. The growing volume of user-generated content on the web has made sentiment analysis (SA) a significant tool for extracting data regarding human emotional states. Twitter sentiment detectors provide a superior solution for assessing the quality of products and services compared to other conventional technologies. The detection performance and classifier accuracy of SA, which can be highly dependent on classifier methods and the quality of input features have been utilised. Deep learning (DL) methods use disti...

Word segmentation is the first and obligatory task for every NLP. For inflectional languages like English, French, Dutch,.. their word boundaries are simply assumed to be whitespaces or punctuations. Whilst in various Asian languages, including Chinese and Vietnamese, ...

For improving the interaction between students and teachers, it is fundamental for teachers to understand students' learning levels. An intelligent computer system should be able to automatically evaluate students' answers when the teacher asks some questions. We first built the assessment corpus from the course in the university. With the corpus, we applied the following procedures to extract the relevant information and then built the feature model: (1) remove the punctuation and decimal numbers because it plays the noise roles, (2) apply the part-of-speech tagging such that the syntactic information is extracted, (3) for grouping the information, take the stemming and normalization procedure to sentences, and (4) extract other features. In this study, we treated the assessment problem as the classifying problem, and tried two kinds of classification strategies: two and three classifying classes. For two classes, we got an average of 66.3% precision rate at first. When add...

Recently, deep learning-based test case generation approaches have been proposed to automate the generation of unit test cases. In this study, we leverage Transformer-based code models to generate unit tests with the help of Domain Adaptation (DA) at a project level. Specifically, we use CodeT5, a relatively small language model trained on source code data, and fine-tune it on the test generation task. Then, we apply domain adaptation to each target project data to learn project-specific knowledge (project-level DA). We use the Methods2test dataset to fine-tune CodeT5 for the test generation task and the Defects4j dataset for project-level domain adaptation and evaluation. We compare our approach with (a) CodeT5 finetuned on the test generation without DA, (b) the A3Test tool, and (c) GPT-4 on five projects from the Defects4j dataset. The results show that tests generated using DA can increase the line coverage by 18.62%, 19.88%, and 18.02% and mutation score by 16.45%, 16.01%, and 12.99% compared to the above (a), (b), and (c) baselines, respectively. The overall results show consistent improvements in metrics such as parse rate, compile rate, BLEU, and CodeBLEU. In addition, we show that our approach can be seen as a complementary solution alongside existing search-based test generation tools such as EvoSuite, to increase the overall coverage and mutation scores with an average of 34.42% and 6.8%, for line coverage and mutation score, respectively. • Software and its engineering → Software testing and debugging; Empirical software validation.

Being successful in intercultural communication relies strongly on understanding of communicative purposes of interlocutors and pragmatic meaning of their utterances. Established on numerous cross-cultural studies, politeness is a culture-specific phenomenon despite its universal nature and understanding of politeness as well as norms of politeness vary across different cultures (

O‗zbekiston Milliy universiteti Jurnalistika fakulteti Kompyuter lingvistikasi yo‗nalishi 1-kurs magistranti Annotatsiya: Ushbu maqolada parallel korpusning matn tarjimasida o‗ziga xos o‗rni hamda tillarni o‗rganish va bir-biriga qiyoslashdagi ahamiyati yoritib berilgan. Shuningdek, parallel korpus tuzish texnologiyasiga ham alohida e'tibor qaratilgan. Kalit so"zlar: mashina tarjimasi, korpus, korpus lingvistikasi, parallel korpus, parallel konkordans, bitex. Annotation: This article will discuss the role of the parallel corpus in text translation and its importance in the study and comparison of languages. Moreover, the technology compiling of parallel corpus was taken into special account.

O‗zbekiston Milliy universiteti Jurnalistika fakulteti Kompyuter lingvistikasi yo‗nalishi 1-kurs magistranti Annotatsiya: Ushbu maqolada parallel korpusning matn tarjimasida o‗ziga xos o‗rni hamda tillarni o‗rganish va bir-biriga qiyoslashdagi ahamiyati yoritib berilgan. Shuningdek, parallel korpus tuzish texnologiyasiga ham alohida e'tibor qaratilgan. Kalit so"zlar: mashina tarjimasi, korpus, korpus lingvistikasi, parallel korpus, parallel konkordans, bitex. Annotation: This article will discuss the role of the parallel corpus in text translation and its importance in the study and comparison of languages. Moreover, the technology compiling of parallel corpus was taken into special account.

BACKGROUND: The integration of the rapidly expanding corpus of information about the genome, transcriptome, and proteome, engendered by powerful technological advances, such as microarrays, and the availability of genomic sequence from multiple species, challenges the grasp and comprehension of the scientific community. Despite the existence of text-mining methods that identify biological relationships based on the textual co-occurrence of gene/protein terms

Text mining studies have started to investigae relations between positive and negative opinions and patients' physical health. Several studies linked the personal lexicon with health and the health-related behavior of the individual. However, few text mining studies were performed to analyze opinions expressed in a large volume of user-written Web content. Our current study focused on performing sentiment analysis on several medical forums dedicated to Hearing Loss (HL). We categorized messages posted on the forums as positive, negative and neutral. Our study had two stages: first, we applied manual annotation of the posts with two annotators and have 82.01% overall agreement with kappa 0.65 and then we applied Machine Learning techniques to classify the posts.

A unit of translation is a source text fragment of any length or nature that piques a translator's interest during translation. Alves and Vale proposed the concept of macro and micro translation units based on pauses and times identified in Translog software while translating a document. The objective of this exploratory study was to describe the characteristics of translation units in relation to the limited capacity of working memory (WM) and the information storage-processing capacity. Four trained translators participated in this study, translating a short text about underground mining. The data collection tools were Translog II and Camtasia software. Findings showed that participants used an analytic and sequential processing method and their limited capacity of WM exhibited a lower processing speed, which had an impact on the length of the macro units of translation (UTs).

This study is the second part of a three-part analysis (if it meets the review requirements) of emotions carried out by written documents. These documents were collected from eight students who took part in a three-day stock market experiment in January 2025. In the first part of this research (1), which was previously published, a lexical approach was used to analyze the words' emotional weight in the passages of the written documents. In this study, we considered all the passages in order to analyze a single written document for each student. To analyze the emotional charges in the documents, three Artificial Intelligences (ChatGPT, Gemini, and DeepSeek) have been used, and six queries have been selected. Using emotional couples, our results suggest different analytical processes depending on the Artificial Intelligence requested and a lack of uniformity in the emotional couples generated, according to the queries selected. At least, Artificial Intelligences are able to identify a strong primary and basic emotional trend but seem to have trouble capturing more nuanced emotional levels in a consistent manner. Using multiple queries does not improve the consistency of the results, and, indeed, the most generic query leads to the most uniform results among the selected artificial intelligences.

The native language of a foreign language learner can have an effect on the errors they make because of similarities or differences between the two languages. In order to provide effective error prediction and correction for nonnative English language learners it is important to identify their specific characteristic error patterns that are influenced by their native language. In this paper, we examine analyzing error detection scores to predict the native language of an English language learner. 15 categories of error detection scores are combined to create an error prediction score vector representation of each sentence. The native language is predicted by training an SVM classifier with the error vectors. The results are compared to an SVM classifier trained with just word representations of the learner writing sentences.

It is important for language learners to determine and reflect on their writing errors in order to overcome weaknesses. Each language learner has their own unique writing error characteristics and therefore has different learning needs. In this paper, we analyze the writing errors of foreign language learners on the language learning SNS website Lang-8 to investigate the characteristics of errors by native language. 142,465 sentences were collected from Lang-8 for analysis. For each native language, the predicted scores of 15 error categories from SVM machine learning models are used as a vector representation of each sentence. These score vectors are then clustered to determine error co-occurrence within the same sentence. The results were then analyzed to determine the error characteristics of different native languages.

This paper explores the idea that intelligence, rather than humanity, serves as the primary force driving evolution, with a particular focus on Artificial General Intelligence (AGI). As AI technology progresses, we must reconsider traditional evolutionary perspectives centered on humans. We propose that intelligence, which initially evolved biologically, now manifests itself in artificial systems. Intelligence transcends being a mere tool; it represents a fundamental element of an ongoing evolutionary process that may operate independently of human influence. Comparing genes with intelligence supports the idea that intelligence could surpass genes or species as the dominant driver of evolution. This shift in thinking requires interdisciplinary collaboration to address ethical concerns, AGI's autonomous evolution, and humanity's role in its development. The emergence of AGI prompts a reevaluation of intelligence and its trajectory, marking a critical point where it may surpass human experience and biological limits, offering new cognitive perspectives.

For many second language learners, learning the target language is supposed to be identical with the mastery of the grammar of that language. When we say "the mastery of the grammar", we refer to the mastery of rules which revolve around the structure of the language. The problem arises at that point, actually. In this paper, this problem is going to be dwelled upon analysing a language item. As a matter of fact, this language item can be regarded as a symbol of the problem mentioned above. This language item is "if clauses". "If clauses" are not so easy to cope with since they require a mastery of some structure accompanied with an appropriate situation. The problem arises here, because a clear-cut distinction between the types and usage are not always possible. We ought not to forget that grammar is not the fundamental of the target language, which consists of the finely tuned structural prescriptions only. Language is meaningful as long as it is contextualized. We cannot take over the use of language if it is presented in isolated exercises having a meaningful context. Comprehensive exercises are essential to teach "if clauses". In conclusion, in this paper I intend to suggest a semantic classification of "if clauses" so as to avoid overgeneralization which is a result of the classification of "if clauses" structurally. Making this suggestion, I would like to support it with examples from the works of literature as well. If the classification of the language item in question is made semantically, the comprehension of the subject "if clauses" will be much better.

For many second language learners, learning the target language is supposed to be identical with the mastery of the grammar of that language. When we say "the mastery of the grammar", we refer to the mastery of rules which revolve around the structure of the language. The problem arises at that point, actually. In this paper, this problem is going to be dwelled upon analysing a language item. As a matter of fact, this language item can be regarded as a symbol of the problem mentioned above. This language item is "if clauses". "If clauses" are not so easy to cope with since they require a mastery of some structure accompanied with an appropriate situation. The problem arises here, because a clear-cut distinction between the types and usage are not always possible. We ought not to forget that grammar is not the fundamental of the target language, which consists of the finely tuned structural prescriptions only. Language is meaningful as long as it is contextualized. We cannot take over the use of language if it is presented in isolated exercises having a meaningful context. Comprehensive exercises are essential to teach "if clauses". In conclusion, in this paper I intend to suggest a semantic classification of "if clauses" so as to avoid overgeneralization which is a result of the classification of "if clauses" structurally. Making this suggestion, I would like to support it with examples from the works of literature as well. If the classification of the language item in question is made semantically, the comprehension of the subject "if clauses" will be much better.

Purpose: Ever increasing penetration of the Internet in our lives has led to an enormous amount of multimedia content generation on the internet. Textual data contributes a major share towards data generated on the world wide web. Understanding people's sentiment is an important aspect of natural language processing, but this opinion can be biased and incorrect, if people use sarcasm while commenting, posting status updates or reviewing any product or a movie. Thus, it is of utmost importance to detect sarcasm correctly and make a correct prediction about the people's intentions. Design/methodology/approach: This study tries to evaluate various machine learning models along with standard and hybrid deep learning models across various standardized datasets. We have performed vectorization of text using word embedding techniques. This has been done to convert the textual data into vectors for analytical purposes. We have used three standardized datasets available in public domain and used three word embeddings i.e Word2Vec, GloVe and fastText to validate the hypothesis. The results were analyzed and conclusions are drawn. The key finding is: the hybrid models that include Bidirectional LongTerm Short Memory (Bi-LSTM) and Convolutional Neural Network (CNN) outperform others conventional machine learning as well as deep learning models across all the datasets considered in this study, making our hypothesis valid. Research limitations: Using the data from different sources and customizing the models according to each dataset, slightly decreases the usability of the technique. But, overall this methodology provides effective measures to identify the presence of sarcasm with a minimum average accuracy of 80% or above for one dataset and better than the current baseline results for the other datasets. The results provide solid insights for the system developers to integrate this model into real-time analysis of any review or comment posted in the public domain. This study has various other practical implications for businesses that depend on user ratings and public opinions. This study also provides a launching platform for various researchers to work on the problem of sarcasm identification in textual data.

Around the world, most of the proposed techniques for the identification of sarcasm either take the utterance in isolation or these methods only perform the categorization of the textual data. Very limited work has been done on how to train or manipulate the various parameters related to textual data so that to improve on the accuracy of the classification method. In this article, we are trying to identify the sarcasm in the textual data using neural networks. We have tried to classify the data using convolutional neural networks (CNN), recurrent neural networks (RNN) and a blend of these techniques to improve accuracy. Our work is not limited to the classification of the sarcastic text, we have also tried to measure the impact of the training data, number of epochs and amount of dropout in the network. The paper also discusses the impact of various embedding on the dataset when converting the same dataset into vectors via different word embeddings. We measured the influence of various parameters on the very large-scale Reddit 1 corpus.

Speech recognition system (ASR) is a technology that allows computers receive the input using the spoken words. This technology requires sample words in the pattern matching process that is stored in the database. There is no reference as the fundamental theory to develop database in ASR. So, the research of database development to optimize the performance of the system is required. Mel-scale frequency cepstral coefficients (MFCCs) is used to extract the characteristics of speech signal and backpropagation neural network in quantized vector is used to evaluate likelihood the maximum log values to the nearest pattern in the database. The results shows the robustness of ASR is optimum using 140 samples of data reference for each word with an average of accuracy is 99.95% and duration process is 27.4 msec. The investigation also reported the gender doesn't have significantly influence to the accuracy. From these results it concluded that the performance of ASR can be increased by optimizing the database.

This paper presents a comparative analysis of two different approaches for clustering textual data from the 20 Newsgroups dataset. The first approach leverages a Large Language Model (LLM) to classify each text into predefined categories using zero-shot classification. The second approach applies to the traditional K-Means clustering algorithm on text embeddings. We evaluate both methods by comparing their predicted clusters against true labels for assessment. For K-Means, we also explore a semi-supervised variant with centroid initialization based on true labels.

Motivation: The most widely used literature search techniques, such as those offered by NCBI's PubMed system, require significant effort on the part of the searcher, and inexperienced searchers do not use these systems as effectively as experienced users. Improved literature search engines can save researchers time and effort by making it easier to locate the most important and relevant literature. Results: We have created and optimized a new, hybrid search system for Medline that takes natural text as input and then delivers results with high precision and recall. The combination of a fast, low-sensitivity weighted keyword-based first pass algorithm to cast a wide net to gather an initial set of literature, followed by a unique sentence-alignment based similarity algorithm to rank order those results was developed that is sensitive, fast and easy to use. Several text similarity search algorithms, both standard and novel, were implemented and tested in order to determine which o...

Parallel and comparable corpora represent a crucial resource for different Natural Language Processing tasks like machine translation, lexical acquisition, and knowledge structuring but are also suitable to be consulted by humans for different purposes, such as linguistic teaching, corpus linguistics, translation studies, lexicography, multilingual information browsing. To enhance their exploitation by human users, specially designed interfaces need to be developed. In this paper we present the design and implementation ...

Milica MIHALJEVIC Zavod za hrvatski jezik HR, Zagreb STRUKTURA RJECNTCKOGA CLANKA U TERMINOLOSKIM RJECNTCIMA Struktura rjecnickoga clanka u terminoloskim rjeenicima razliCita je ovisno 0 vrsti i namjeni rjeenika. Neki mali, prijevodni terminoloski rjeenici imaju sa svim jednostavnu strukturu Medutim, u objasnidbenim rjeenicima rjeenicki clanak moze biti vrlo kompleksan. U radu se na primjeru nekih postojecih hrvatskih terminoloskih rjecnika analizira struktura rjeenickoga clanka te pred laze najprihvatljivija struktura za razl icite tipove terminoloskih rjeenika. Terminoloski su rjecnici specijalni rjecnici koji se ovisno 0 klasifikacijskom kri teriju dijele na vise tipova. Budu6i da struktura r jeenickoga clanka u terminoloskom rjeCniku ovisi 0 tipu rjecnika, ovdje cu ukratko nesto re6i 0 podjeli terminoloskih rjeenika. Govorit cu samo 0 onim kriterijima podjele 0 kojima ovisi struktura rje6 nickoga clanka. S obzirom na nacin obradbe natuknice rjecnici se dijele na objas nidbene i suodnosne (korelacijske). Objasnidbeni rjecnici navode defmiciju, katkada primjer uporabe ili ilustraciju. Suodnosni rjecnici samo povezuju (koreliraju) ozna Citelje na istom (sinonimijski rjecnici) ili raznim jezicima (prijevodni rjecnici). Iako je zamisliv terminoloski sinonimijski rjecnik, u praksi su terminoloski rjeenici uglavnom samo prijevodni, objasnidbeni ili prijevodni i objasnidbeni. S obzirom na broj jezika koji su njima obuhvaceni rjecnike dijelimo na jednojezicne i visejezicne (dvojezicne, trojezicne, cetverojezicne, petojezicne itd.). Po opsegu se rjecnici dijele na velike, srednje i male. Terminoloski se rjeCnici obzirom na struku dijele na ma tematicke, kemijske, bioloske, lingvisticke itd. Najjednostavniju strukturu rjecnickoga clanka imaju mali, prijevodni, dvojezicni rjecnici. Oni uz natuknicu l navode istoznacnice na jednom ili vise jezika. Primjer Svaki naziv ima dvije slrane: pojam koji oznacuje i jezicni izraz k<~im se laj pojam ozoaeuje. U rjeeniku se laj jezieni izraz naziva natuknica. Natuknica (nj. Schlagworl. Stichworl. eng!. entry) jest leksicka jedinica (skup rijeCi. rijeC. morfem) koja sluzi kao naslov rjeCnickoga clanka u kojem se opisuje njezino znaeenje. objaSnjava pojam koji je njome oznaeen. Obicno je naluknica istaknuta posebnOO1 vrslom pisma ili slova (najceliCe masnim slovima).