Hasil untuk "English language"

R. M. Dauer

D. Crystal, Derek Davy

Gillian R. Brown, G. Yule, Neil McKelvie

Tamar H. Gollan, Rosa I. Montoya, Cynthia Cera et al.

P. Jusczyk, A. Friederici, J. Wessels et al.

P. Lightbown, N. Spada

M. Cettolo, J. Niehues, S. Stüker et al.

Lydia White

Kanavillil Rajagopala

Samira Al Hosni

Lauren Levine, Junghyun Min, Amir Zeldes

In this paper we present a sample treebank for Old English based on the UD Cairo sentences, collected and annotated as part of a classroom curriculum in Historical Linguistics. To collect the data, a sample of 20 sentences illustrating a range of syntactic constructions in the world's languages, we employ a combination of LLM prompting and searches in authentic Old English data. For annotation we assigned sentences to multiple students with limited prior exposure to UD, whose annotations we compare and adjudicate. Our results suggest that while current LLM outputs in Old English do not reflect authentic syntax, this can be mitigated by post-editing, and that although beginner annotators do not possess enough background to complete the task perfectly, taken together they can produce good results and learn from the experience. We also conduct preliminary parsing experiments using Modern English training data, and find that although performance on Old English is poor, parsing on annotated features (lemma, hyperlemma, gloss) leads to improved performance.

Muhamad Adnan Royandi, Enung Hasanah, Suyatno Suyatno

The ever-growing era of globalization places demands on having 21st century skills which include Critical Thinking, Creativity, Communication, and Collaboration. Technological advances require us to have communication and English language skills, and as Muslims, we are also required to have Arabic language skills to understand the holy book. In its implementation, Student Management has the function of improving these abilities and skills. This research aims to determine the planning, organization, implementation and control, and supervision of boarding school student management in improving Arabic and English language skills. This research uses a qualitative type of research, with a field observation approach, and data collection techniques using interviews, observation, and documentation. The results of this research explain that boarding school student management in improving Arabic and English language skills is by: 1) Planning which begins with a Work Meeting which plans one semester's work program, committee design, budget, and activity calendar; 2) Organizing by dividing tasks, committee obligations, and coordinating with PLC, LA, and SOS; 3) Implementation of daily language programs using Arabic and English which change every 2 weeks, weekly programs with vocabulary learning, public speaking, and muhadatsah, as well as semester programs in the form of language competitions such as the LHI Language Competition and Public Speaking Contest; 4) Control and supervision carried out by all teachers and also dormitory supervisors as well as evaluations held every week.

Daniel Bekele Wakjira, Abdi Deksisa, Wondu Abera et al.

Abstract Background Yoga and Reflexology is a non-pharmacological pain relief method, the effects of which remain unclear on obstetric interventions. This systematic review and meta-analysis aimed to assess the effect of yoga and reflexology on the length of labor and its pain. Methods In this review, PubMed, Medline, Scopus, Cochrane, PsycINFO, and Google scholar databases were searched from inception till April 22, 2024. The protocol registry number (ID: CRD 42022362769) assigned. English-language of studies in randomized controlled trials (RCTs were included in the review. The Cochrane risk of bias tool was used to assess the risk of bias. Two independent reviewers extracted data and assessed the quality of the trials. Using Rev Man 5.4 software, the results were presented in form of forest plots. Results Ten studies including 1,253 pregnant women were included for meta-analysis. Prenatal yoga and reflexology interventions has significant effect in reducing labor duration at fist stage (SMD: −0.43; 95% CI: −0.74, −0.13; I2 = 77%), second stage (SMD: −0.34; 95% CI: −0.53, −0.15; I2 = 21%) and third stage (SMD: −1.05; 95% CI: −1.34, −0.76; I2 = 0%, P = 0.0001). Moreover, the labor pain perceived in the active phase was found to be significantly reduced in the intervention group (SMD: −1.56; 95% CI: −2.78, −0.34; P = 0.0001). Conclusion The review has demonstrated that yoga and reflexology is a non-invasive health exercises during prenatal period which is useful in reducing labor pain and its length at each stage of labor. Therefore, through applying this technique, a goal of midwifery, which is reducing labor pain and its length, can be achieved. It also may enable moms to fully enjoy their pregnancy and delivery.

Marian W. Makins

Oline Ranum, Gomer Otterspeer, Jari I. Andersen et al.

In this work, we present an efficient approach for capturing sign language in 3D, introduce the 3D-LEX v1.0 dataset, and detail a method for semi-automatic annotation of phonetic properties. Our procedure integrates three motion capture techniques encompassing high-resolution 3D poses, 3D handshapes, and depth-aware facial features, and attains an average sampling rate of one sign every 10 seconds. This includes the time for presenting a sign example, performing and recording the sign, and archiving the capture. The 3D-LEX dataset includes 1,000 signs from American Sign Language and an additional 1,000 signs from the Sign Language of the Netherlands. We showcase the dataset utility by presenting a simple method for generating handshape annotations directly from 3D-LEX. We produce handshape labels for 1,000 signs from American Sign Language and evaluate the labels in a sign recognition task. The labels enhance gloss recognition accuracy by 5% over using no handshape annotations, and by 1% over expert annotations. Our motion capture data supports in-depth analysis of sign features and facilitates the generation of 2D projections from any viewpoint. The 3D-LEX collection has been aligned with existing sign language benchmarks and linguistic resources, to support studies in 3D-aware sign language processing.

Brendon Boldt, David Mortensen

In this paper, we introduce a benchmark for evaluating the overall quality of emergent languages using data-driven methods. Specifically, we interpret the notion of the "quality" of an emergent language as its similarity to human language within a deep learning framework. We measure this by using the emergent language as pretraining data for a downstream NLP tasks in human language -- the better the downstream performance, the better the emergent language. We implement this benchmark as an easy-to-use Python package that only requires a text file of utterances from the emergent language to be evaluated. Finally, we empirically test the benchmark's validity using human, synthetic, and emergent language baselines.

Vincent Beliveau, Helene Kaas, Martin Prener et al.

Natural language processing (NLP) in the medical domain can underperform in real-world applications involving small datasets in a non-English language with few labeled samples and imbalanced classes. There is yet no consensus on how to approach this problem. We evaluated a set of NLP models including BERT-like transformers, few-shot learning with sentence transformers (SetFit), and prompted large language models (LLM), using three datasets of radiology reports on magnetic resonance images of epilepsy patients in Danish, a low-resource language. Our results indicate that BERT-like models pretrained in the target domain of radiology reports currently offer the optimal performances for this scenario. Notably, the SetFit and LLM models underperformed compared to BERT-like models, with LLM performing the worst. Importantly, none of the models investigated was sufficiently accurate to allow for text classification without any supervision. However, they show potential for data filtering, which could reduce the amount of manual labeling required.

Manase H Halitopo, Napius Kogoya

Oral literature is a cultural expression that verbally transmits values from one generation to the next among the community supporters. This oral literature research aims at (1) presenting the stories' plots through the research's primary goal, 2) introducing the characters that appear in the story, (3) outlining the characters in the stories, (4) recognizing and explaining the settings of the stories, and 5) recognizing and defining symbols in the stories based on Yektingtyas's points of view (2016). The approach in this research incorporates literary criticism into secondary and primary qualitative data combined throughout the data collection process. The primary data were analyzed thoroughly, while the secondary data were utilized as references to analyze the primary data. The results show that Ilik was born to a mother under the shade without a father. Initially, the narrative focused solely on the character's day-to-day activities. The folks washing the pig gut rope cast away and flung the dog with the stone when he tried to devour the remaining portion. The pregnant woman and her children arrived in the village the following day to cook the raw pork they had managed to rescue. The portrait of this literary work depicts the characters’ lives and social harmony, allowing them to nurture indigenous values while learning to explore more life lessons from society.  

G. Valdés

Alex Mei, Sharon Levy, William Yang Wang

As large language models are integrated into society, robustness toward a suite of prompts is increasingly important to maintain reliability in a high-variance environment.Robustness evaluations must comprehensively encapsulate the various settings in which a user may invoke an intelligent system. This paper proposes ASSERT, Automated Safety Scenario Red Teaming, consisting of three methods -- semantically aligned augmentation, target bootstrapping, and adversarial knowledge injection. For robust safety evaluation, we apply these methods in the critical domain of AI safety to algorithmically generate a test suite of prompts covering diverse robustness settings -- semantic equivalence, related scenarios, and adversarial. We partition our prompts into four safety domains for a fine-grained analysis of how the domain affects model performance. Despite dedicated safeguards in existing state-of-the-art models, we find statistically significant performance differences of up to 11% in absolute classification accuracy among semantically related scenarios and error rates of up to 19% absolute error in zero-shot adversarial settings, raising concerns for users' physical safety.