Hasil untuk "Computational linguistics. Natural language processing"

Wuzhen Shi, Youwei Pan, Chun Zhao et al.

ABSTRACT Although Transformer‐based image restoration methods have demonstrated impressive performance, existing Transformers still insufficiently exploit multiscale information. Previous non‐Transformer‐based studies have shown that incorporating multiscale features is crucial for improving restoration results. In this paper, we propose a multiscale Transformer (MST) that captures cross‐scale attention among tokens, thereby effectively leveraging the multiscale patch recurrence prior of natural images. Furthermore, we introduce a channel‐gate feed‐forward network (CGFN) to enhance inter‐channel information aggregation and reduce channel redundancy. To simultaneously utilise global, local and multiscale features, we design a multitype feature integration block (MFIB). Extensive experiments on both image super‐resolution and HEVC compressed video artefact reduction demonstrate that the proposed MST achieves state‐of‐the‐art performance. Ablation studies further verify the effectiveness of each proposed module.

BAHATI MITIMA Emmanuel

Résumé : Cette étude examine l’influence des réseaux sociaux sur les violences urbaines en RDC, dans un contexte marqué par la précarité, l’urbanisation rapide et l’exclusion des jeunes. Basée sur une approche qualitative (entretiens, analyses documentaires et numériques), elle mobilise la théorie de la frustration relative et celle de la spirale médiatique pour analyser les dynamiques de tensions. Elle innove en abordant les violences urbaines numériques peu explorées dans la littérature et met en lumière le rôle des influenceurs et groupes communautaires en ligne. L’étude propose une typologie des violences numériques et recommande une régulation concertée entre État, plateformes et société civile. Les réseaux sociaux sont ainsi présentés comme des catalyseurs ambivalents : à la fois outils de mobilisation citoyenne et facteurs de polarisation. Mots clés : Violences Urbaines, Réseaux sociaux, Médias sociaux, Fake News, Sécurité.

Helia Farhood, Magnus Nyden, Amin Beheshti et al.

Abstract Personalised learning models can assist students in meeting their unique requirements and objectives for expanding their knowledge, perspective, abilities, and understanding of the educational system. With the advancement of artificial intelligence, technological integration has the potential to play a critical role in personalising the learning experience. This work attempts a systematic review that examines the influence of personalised learning enabled by artificial intelligence on the educational system. The review synthesises the latest research literature (January 2015–June 2025) and includes a total of 125 studies that achieved our inclusion criteria to demonstrate how technology could successfully modify the learning system. Thoroughly examining the role of artificial intelligence–based personalised learning in education is the main focus of this study. In particular, we categorise applications and review algorithms that support personalised learning, compare their approaches, and identify their impact on teaching, learning, and assessment practices across education sectors (K–12, higher education, and institutionally supported online learning). This work contributes to the field of education by providing a much-needed summary of the current state of research and the various opportunities and challenges in this area.

Yana Zhou

Abstract Accurate traffic flow prediction is a crucial aspect of intelligent transportation systems, yet it remains challenging due to disruptions caused by non-routine events such as accidents, road closures, and severe weather. Traditional statistical and machine learning models often fail to fully capture these anomalies, resulting in reduced prediction accuracy. This research proposes a hybrid approach that enhances traffic forecasting by integrating social media data, specifically tweets, to detect real-time events that may impact traffic conditions. The methodology consists of three key phases. First, social media data is collected and preprocessed to remove noise, spam, and irrelevant content, thereby reducing data complexity without sacrificing information quality. Second, a Hidden Markov model is applied to analyze time-series patterns involving tweet frequency, temporal factors, and weather conditions. Finally, a SincNet-based convolutional neural network is utilized for high-accuracy traffic prediction. SincNet’s ability to effectively process time-series data makes it particularly well-suited for modeling traffic behavior. The model demonstrates strong performance in forecasting morning traffic based on tweet data available up to midnight of the previous day. Experimental results show that the proposed approach achieves improved prediction accuracy of up to 90%, with 89% precision and a reduced error rate of 0.2, outperforming existing state-of-the-art models.

BECHIRI Camélia & BENACHOUR Yamina

Résumé : L’article explore le rôle de l’erreur dans les processus d’apprentissage, en s’appuyant sur les avancées des neurosciences cognitives pour proposer une réhabilitation de l’erreur en pédagogie. Traditionnellement perçue comme un échec ou une faute à éviter, l’erreur est ici présentée comme un mécanisme essentiel du fonctionnement cérébral, favorisant la plasticité neuronale et la consolidation des savoirs. Le cerveau, en tant que machine prédictive, utilise les erreurs pour ajuster ses modèles mentaux et améliorer ses prédictions. L’article retrace l’évolution historique de la perception de l’erreur, de l’Antiquité à l’époque moderne, en montrant comment elle a oscillé entre une vision exploratoire (chez Socrate ou Aristote) et une connotation répressive (au Moyen Âge). Il analyse également les modèles pédagogiques, du modèle transmissif où l’erreur est une faute au constructivisme où elle devient un obstacle épistémologique à dépasser, en passant par le béhaviorisme qui cherche à minimiser les erreurs. Les neurosciences cognitives apportent un éclairage nouveau, en considérant l’erreur comme un signal d’apprentissage essentiel. Des stratégies pédagogiques inspirées par ces recherches, comme l’effet de test, l’apprentissage par essai-erreur et les feedbacks explicites, montrent que l’erreur, lorsqu’elle est analysée et corrigée de manière constructive, devient un puissant levier pour renforcer la mémorisation et la compréhension. Une expérimentation menée auprès d’étudiants en grammaire sur l’accord du COD illustre ces principes. En utilisant des QCM et en analysant collectivement les erreurs, les étudiants ont montré une nette amélioration de leurs performances, confirmant que l’erreur, loin d’être un obstacle, peut devenir un outil d’apprentissage efficace. En conclusion, l’article plaide pour une pédagogie bienveillante qui intègre l’erreur comme une étape naturelle et constructive du processus d’apprentissage, en s’appuyant sur les enseignements des neurosciences pour transformer les difficultés des élèves en opportunités de progression. Mots-clés : erreur –apprentissage –neurosciences cognitives –pédagogie-QCM

Han Lin, Jiatong Han, Pingping Wu et al.

Abstract As human‐machine interaction (HMI) in healthcare continues to evolve, the issue of trust in HMI in healthcare has been raised and explored. It is critical for the development and safety of healthcare that humans have proper trust in medical machines. Intelligent machines that have applied machine learning (ML) technologies continue to penetrate deeper into the medical environment, which also places higher demands on intelligent healthcare. In order to make machines play a role in HMI in healthcare more effectively and make human‐machine cooperation more harmonious, the authors need to build good human‐machine trust (HMT) in healthcare. This article provides a systematic overview of the prominent research on ML and HMT in healthcare. In addition, this study explores and analyses ML and three important factors that influence HMT in healthcare, and then proposes a HMT model in healthcare. Finally, general trends are summarised and issues to consider addressing in future research on HMT in healthcare are identified.

Cuijin Li, Zhong Qu, Shengye Wang

Abstract Automatically detecting and locating remote occlusion small objects from the images of complex traffic environments is a valuable and challenging research. Since the boundary box location is not sufficiently accurate and it is difficult to distinguish overlapping and occluded objects, the authors propose a network model with a second‐order term attention mechanism and occlusion loss. First, the backbone network is built on CSPDarkNet53. Then a method is designed for the feature extraction network based on an item‐wise attention mechanism, which uses the filtered weighted feature vector to replace the original residual fusion and adds a second‐order term to reduce the information loss in the process of fusion and accelerate the convergence of the model. Finally, an objected occlusion regression loss function is studied to reduce the problems of missed detections caused by dense objects. Sufficient experimental results demonstrate that the authors’ method achieved state‐of‐the‐art performance without reducing the detection speed. The mAP@.5 of the method is 85.8% on the Foggy_cityscapes dataset and the mAP@.5 of the method is 97.8% on the KITTI dataset.

Maminiaina René Alexandre TSISAROTINA & Hery Zo ANDRIAMANOHISOA

Résumé : L'entrepreneuriat cognitif, caractérisé par l'utilisation des capacités cognitives humaines pour innover et créer de la valeur, représente une approche novatrice dans le domaine de l'entrepreneuriat. Cette approche intègre les principes des sciences cognitives dans le processus de création et de gestion d'entreprise, offrant ainsi de nouvelles perspectives et opportunités pour réussir dans un monde des affaires en constante évolution. L'optimisation de l'entrepreneuriat cognitif implique la maximisation des ressources mentales et intellectuelles des entrepreneurs, nécessitant une compréhension approfondie des mécanismes cognitifs sous-jacents, tels que la pensée critique, la résolution de problèmes, la créativité et la prise de décision. La valorisation de l'entrepreneuriat cognitif consiste à reconnaître l'importance des compétences cognitives dans le succès entrepreneurial, promouvant une culture d'entreprise valorisant la pensée critique, la résolution de problèmes, la créativité et l'apprentissage continu. L'entrepreneuriat cognitif joue également un rôle crucial dans la transformation des industries et des marchés existants en introduisant des approches novatrices basées sur les sciences cognitives, perturbant les modèles d'affaires traditionnels et redéfinissant l'interaction avec les clients. L'optimisation et la valorisation de l'entrepreneuriat cognitif revêtent une importance stratégique pour faire face à des défis complexes tels que la transition vers une économie numérique, la gestion de l'incertitude économique et environnementale, et la promotion de l'innovation sociale. Mots clés : Processus de création, Entrepreneuriat cognitif, capacités cognitives humaines, Innovation, Création de valeur, Approche novatrice, Optimisation des ressources mentales.

Yuxia Wang, Shimin Tao, Ning Xie et al.

Dr. Mirza Hamid Baig

This article explores the intersection between the Progressive poetic tradition and the ghazal. The Progressive movement, known as "Taraqi Passand" in Urdu, emerged as a literary movement in the 20th century, aiming to challenge conventional poetic norms and embrace new ideas and social realities. While the ghazal traditionally revolved around themes of love, longing, and beauty, the Progressive poets sought to expand its boundaries and infuse it with socio-political relevance. This article delves into the historical context of the Progressive poetic tradition, highlighting its goals of addressing social injustices, advocating for political reform, and amplifying the voices of marginalized communities. By incorporating these concerns into the ghazal, the Progressive poets transformed the traditional form into a powerful tool for social commentary and critique. Furthermore, this article discusses the thematic evolution of the ghazal within the Progressive tradition. It explores how the poets expanded the traditional themes to encompass issues of inequality, poverty, discrimination, and the challenges of modernization. By doing so, the Progressive poets widened the scope of the ghazal and made it a reflection of the changing times and the evolving concerns of society. Through an exploration of the blending of the ghazal's structural framework with progressive ideas, this article emphasizes the significant role played by the Progressive poets in shaping the modern Urdu literary landscape. By infusing the ghazal with socio-political perspectives, they brought depth, diversity, and relevance to the traditional form, and in turn, opened new possibilities for poetic expression. In summary, this article highlights the symbiotic relationship between the Progressive poetic tradition and the ghazal, showcasing how the poets of this movement used the form to convey their socio-political concerns and contribute to the broader discourse of their time.

Rialda Spahic, Mary Ann Lundteigen, Vidar Hepsø

Abstract This research examines the factors contributing to the exterior material degradation of subsea oil and gas pipelines monitored with autonomous underwater systems (AUS). The AUS have a role of gathering image data that is further analyzed with artificial intelligence data analysis methods. Corrosion and potential ruptures on pipeline surfaces are complex processes involving several competing elements, such as the geographical properties, composition of soil, atmosphere, and marine life, whose eflt in substantial environmental damage and financial loss. Despite extensive research, corrosion monitoring and prediction remain a persistent challenge in the industry. There is a lack of knowledge map that can enable image ausing an AUS to recognize ongoing degradation processes and potentially prevent substantial damage. The main contribution of this research is the knowledge map for increased context and risk awareness to improve the reliability of image-based monitoring and inspection by autonomous underwater systems in detecting hazards and early signs of material degradation on subsea pipeline surfaces.

Olga Majewska, A. Korhonen

Recent developments in language modeling have enabled large text encoders to derive a wealth of linguistic information from raw text corpora without supervision. Their success across natural language processing (NLP) tasks has called into question the role of man-made computational resources, such as verb lexicons, in supporting modern NLP. Still, probing analyses have concurrently exposed the limitations of the knowledge possessed by the large neural architectures, revealing them to be clever task solvers rather than self-taught linguists. Can human-designed lexical resources still help fill their knowledge gaps? Focusing on verb classification, we discuss approaches to generating verb classes multilingually and weigh the relative benefits of undertaking expensive lexicographic work and outsourcing the task to untrained native speakers. Then, we consider the evidence for the utility of augmenting pretrained language models with external verb knowledge and ponder the ways in which human expertise can continue to benefit multilingual NLP. Expected final online publication date for the Annual Review of Linguistics, Volume 9 is January 2023. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Yingxu Wang

The fuzzy properties of language semantics are a central problem towards machine-enabled natural language processing in cognitive linguistics, fuzzy systems, and computational linguistics. A formal method for rigorously describing and manipulating fuzzy semantics is sought for bridging the gap between humans and cognitive fuzzy systems. The mathematical model of fuzzy concepts is rigorously described as a hyperstructure of fuzzy sets of attributes, objects, relations, and qualifications, which serves as the basic unit of fuzzy semantics for denoting languages entities in semantic analyses. The formal fuzzy concept is extended to complex structures where fuzzy modifiers and qualifiers are considered. An algebraic approach is developed to manipulate composite fuzzy semantic as a deductive process from a fuzzy concept to the determined semantics. The denotational mathematical structure of fuzzy semantic inference not only explains the fuzzy nature of human semantics and its comprehension, but also enables cognitive machines and fuzzy systems to mimic the human fuzzy inference mechanisms in cognitive linguistics, cognitive computing, and computational intelligence.

Rima Sarah, Agry Pramita

Culture and phenomena that exist in society can be known through novels. One of them is the phenomenon of women. This study discusses the woman struggle through two works of English and Indonesian literature, namely the novel Jane Eyre by Charlotte Bronte and the novel Entrok by Okky Madasari. These two novels tell about the life of the women character and their environment. This study uses the theory of feminist literary criticism to find feminist issues in the two novels that appear through the female characters. The method used is descriptive analysis using a qualitative approach. The data source is the novel Jane Eyre by Charlotte Bronte, published in 1847 by Smith, Elder & Co. in London, England, and Okky Madasari's novel Entrok published by Gramedia Pustaka Utama in 2010. The results of data analysis show that the female characters, namely Marni, Rahayu, and Jane Eyre have similarities in making efforts to equalize women through freedom of choice, education. , and work. However, there are also differences regarding the setting of place, time, and culture in the two novels.

Nur Lan, Michal Geyer, Emmanuel Chemla et al.

Vihaan Nama, Dr. Vinay V Hegde, D. B. S. Babu

Movie reviews are vital in telling the viewer whether a movie is worth watching or not. They can be classified into textual and non-textual movie reviews. While non-textual movie reviews (stars) give the user information as to how the movie fairs, textual movie reviews give the user a more detailed picture on the positive and negative aspects of the movie. Sentiment Analysis is the use of natural language processing, text analysis, biometrics and computational linguistics to identify, quantify, extract and effectively study states and subjective information given in textual format. This paper aims to conduct sentiment analysis of reviews of movies by using the Naive-Bayes algorithm and compare the results to that of a Rule-Based Approach using the AFINN-111 sentiment dictionary.

Shahnaz Akhtar

<p class="MsoNoSpacing" style="mso-pagination: none;"><span style="font-size: 12.0pt;">Man is superior to other living beings because he uses language and expresses his views through an artistic way. Poetry is very natural among other genres of literature. Poetry is a combination of different forms of art like music, sculpture, dance and painting etc. Poetry is created through imagination and scenses. Modern urdu poetry especially modern urdu poem and other genres have been introduced under the impact and influence of English literature, Women&rsquo;s contribution in poetry is a clear sign of new and different approaches in relation between women writers and literature. In this article female&rsquo;s contribution in jadeed Nazm (free verse) has been highlighted especially. The well known poetess &ldquo;Nigah Zahra&rdquo; has been discussed, because she has been contributed creative work especially in poetry by visual images and senses and present a descriptive work in free verse by using five senses.</span></p>

Katja Markert, Malvina Nissim

Maryna Bogachyk, Dmytro Bihunov

The Structural-Semantic Features of Computer Terms in English The article is devoted to the study of the structural and semantic characteristics of computer terms in English. The peculiarities of the word-formation process and the functioning of English computer terms are analysed.   Strukturalne i semantyczne cechy terminów komputerowych w języku angielskim Artykuł poświęcony jest badaniu cech strukturalnych i semantycznych terminów komputerowych w języku angielskim. Przeanalizowano specyfikę procesu słowotwórstwa i funkcjonowania angielskich terminów komputerowych.

Adam Tsakalidis, Marya Bazzi, Mihai Cucuringu et al.

Semantic change detection (i.e., identifying words whose meaning has changed over time) started emerging as a growing area of research over the past decade, with important downstream applications in natural language processing, historical linguistics and computational social science. However, several obstacles make progress in the domain slow and difficult. These pertain primarily to the lack of well-established gold standard datasets, resources to study the problem at a fine-grained temporal resolution, and quantitative evaluation approaches. In this work, we aim to mitigate these issues by (a) releasing a new labelled dataset of more than 47K word vectors trained on the UK Web Archive over a short time-frame (2000-2013); (b) proposing a variant of Procrustes alignment to detect words that have undergone semantic shift; and (c) introducing a rank-based approach for evaluation purposes. Through extensive numerical experiments and validation, we illustrate the effectiveness of our approach against competitive baselines. Finally, we also make our resources publicly available to further enable research in the domain.