English Datasets Research Articles

Risk of serious adverse events after primary shoulder replacement: development and external validation of a prediction model using linked national data from England and Denmark

Despite a rising rate of serious medical complications after shoulder replacement surgery, there are no prediction models in widespread use to guide surgeons in identifying patients at high risk and to provide patients with personalised risk estimates to support shared decision making. Our aim was to develop and externally validate a prediction model for serious adverse events within 90 days of primary shoulder replacement surgery. Linked data from the National Joint Registry, National Health Service Hospital Episode Statistics Admitted Patient Care of England, and Civil Registration Mortality databases and Danish Shoulder Arthroplasty Registry and National Patient Register were used for our modelling study. Patients aged 18-100 years who had a primary shoulder replacement between April 1, 2012, and Oct 2, 2020, in England, and April 1, 2012, and Oct 2, 2018, in Denmark, were included. We developed a multivariable logistic regression model using the English dataset to predict the risk of 90-day serious adverse events, which were defined as medical complications requiring admission to hospital and all-cause death. We undertook internal validation using bootstrapping, and internal-external cross-validation across different geographical regions of England. The English model was externally validated on the Danish dataset. Data for 40 631 patients undergoing primary shoulder replacement (mean age 72·5 years [SD 9·9]; 28 709 [70·7%] women and 11 922 [29·3%] men) were used for model development, of whom 2270 (5·6%) had a 90-day serious adverse event. On internal validation, the model had a C-statistic of 0·717 (95% CI 0·707-0·728) and was well calibrated. Internal-external cross-validation showed consistent model performance across all regions in England. Upon external validation on the Danish dataset (n=6653; mean age 70·5 years [SD 10·3]; 4503 [67·7%] women and 2150 [32·3%] men), the model had a C-statistic of 0·750 (95% CI 0·723-0·776). Decision curve analysis showed clinical utility, with net benefit across all risk thresholds. This externally validated prediction model uses commonly available clinical variables to accurately predict the risk of serious medical complications after primary shoulder replacement surgery. The model is generalisable and applicable to most patients in need of a shoulder replacement. Its use offers support to clinicians and could inform and empower patients in the shared decision-making process. National Institute for Health and Care Research and the Department of Orthopaedic Surgery, Herlev and Gentofte Hospital, Denmark.

Advanced differential evolution for gender-aware English speech emotion recognition

Speech emotion recognition (SER) technology involves feature extraction and prediction models. However, recognition efficiency tends to decrease because of gender differences and the large number of extracted features. Consequently, this paper introduces a SER system based on gender. First, gender and emotion features are extracted from speech signals to develop gender recognition and emotion classification models. Second, according to gender differences, distinct emotion recognition models are established for male and female speakers. The gender of speakers is determined before executing the corresponding emotion model. Third, the accuracy of these emotion models is enhanced by utilizing an advanced differential evolution algorithm (ADE) to select optimal features. ADE incorporates new difference vectors, mutation operators, and position learning, which effectively balance global and local searches. A new position repairing method is proposed to address gender differences. Finally, experiments on four English datasets demonstrate that ADE is superior to comparison algorithms in recognition accuracy, recall, precision, F1-score, the number of used features and execution time. The findings highlight the significance of gender in refining emotion models, while mel-frequency cepstral coefficients are important factors in gender differences.

A Survey on Image Caption Generation in Various Languages

The image caption is the process of adding an explicit, coherent description to the contents of the image. This is done by using the latest deep learning techniques, which include computer vision and natural language processing, to understand the contents of the image and give it an appropriate caption. Multiple datasets suitable for many applications have been proposed. The biggest challenge for researchers with natural language processing is that the datasets are incompatible with all languages. The researchers worked on translating the most famous English data sets with Google Translate to understand the content of the images in their mother tongue. In this paper, the proposed review aims to enhance the understanding of image captioning strategies and to survey previous research related to image captioning while examining the most popular databases in different languages, mostly English, translating into other languages using the latest models for describing images, summarizing evaluation measures, and comparing them.

Multimodal Affective Communication Analysis: Fusing Speech Emotion and Text Sentiment Using Machine Learning

Affective communication, encompassing verbal and non-verbal cues, is crucial for understanding human interactions. This study introduces a novel framework for enhancing emotional understanding by fusing speech emotion recognition (SER) and sentiment analysis (SA). We leverage diverse features and both classical and deep learning models, including Gaussian naive Bayes (GNB), support vector machines (SVMs), random forests (RFs), multilayer perceptron (MLP), and a 1D convolutional neural network (1D-CNN), to accurately discern and categorize emotions in speech. We further extract text sentiment from speech-to-text conversion, analyzing it using pre-trained models like bidirectional encoder representations from transformers (BERT), generative pre-trained transformer 2 (GPT-2), and logistic regression (LR). To improve individual model performance for both SER and SA, we employ an extended dynamic Bayesian mixture model (DBMM) ensemble classifier. Our most significant contribution is the development of a novel two-layered DBMM (2L-DBMM) for multimodal fusion. This model effectively integrates speech emotion and text sentiment, enabling the classification of more nuanced, second-level emotional states. Evaluating our framework on the EmoUERJ (Portuguese) and ESD (English) datasets, the extended DBMM achieves accuracy rates of 96% and 98% for SER, 85% and 95% for SA, and 96% and 98% for combined emotion classification using the 2L-DBMM, respectively. Our findings demonstrate the superior performance of the extended DBMM for individual modalities compared to individual classifiers and the 2L-DBMM for merging different modalities, highlighting the value of ensemble methods and multimodal fusion in affective communication analysis. The results underscore the potential of our approach in enhancing emotional understanding with broad applications in fields like mental health assessment, human–robot interaction, and cross-cultural communication.

Trends in incidence of pneumothorax in England before, during and after the COVID-19 pandemic (2017–2023): a population-based observational study

COVID-19 is a risk factor for pneumothorax. The pandemic may have influenced healthcare-seeking behaviour for pneumothorax. This study aimed to investigate recent trends in the incidence of pneumothorax in England. A population-based epidemiological study was conducted using an English national dataset of hospital admissions (Hospital Episode Statistics) from 2017 to 2023. Record-linkage was used to identify multiple admissions per person and co-morbidity. Pneumothoraces co-occurring with COVID-19 were identified by concurrent COVID-19 diagnostic coding. The pre-pandemic (January 2017-February 2020), pandemic (March-2020-February-2021) and post-pandemic periods (March 2021-March 2023) were compared. From 2017 to 2023, there were 72,275 hospital admissions for spontaneous pneumothorax among 59,130 patients. Admissions showed marked variability, peaking in January 2021 when the rate of admissions was about two-thirds higher than that of the pre-pandemic level (Incidence rate ratio [IRR] 1.65, 95% CI: 1.48-1.84). However, when excluding patients with a concurrent COVID-19 diagnosis, the overall trend shifted to a reduction during the pandemic period. Post-pandemic rates were not significantly different from pre-pandemic levels (IRR=0.96, 95% CI: 0.89-1.04). The incidence of spontaneous pneumothorax was significantly higher in males (rate ratio compared to females: 2.29, 95% CI: 2.19-2.39). However, the trends were consistent in both males and females. This study highlights a significant peak in COVID-19 related cases but a corresponding trough in non-COVID-related cases (end 2020, early 2021). Despite a previous report of increasing incidence of (non-COVID-related) hospitalised spontaneous pneumothorax over the long-term between 1968 and 2016, we did not observe any continued increase throughout this study period, prompting further investigation into the impact of recent guidelines. Authors are supported by the NIHR Oxford BRC, Li Ka Shing and Robertson Foundations, MRC, and HDR UK.

Applied Hedge Algebra Approach with Multilingual Large Language Models to Extract Hidden Rules in Datasets for Improvement of Generative AI Applications

Generative AI applications have played an increasingly significant role in real-time tracking applications in many domains including, for example, healthcare, consultancy, dialog boxes (common types of window in a graphical user interface of operating systems), monitoring systems, and emergency response. This paper considers generative AI and presents an approach which combines hedge algebra and a multilingual large language model to find hidden rules in big data for ChatGPT. We present a novel method for extracting natural language knowledge from large datasets by leveraging fuzzy sets and hedge algebra to extract these rules, presented in meta data for ChatGPT and generative AI applications. The proposed model has been developed to minimize the computational and staff costs for medium-sized enterprises which are typically resource and time limited. The proposed model has been designed to automate question–response interactions for rules extracted from large data in a multiplicity of domains. The experimental results show that the proposed model performs well using datasets associated with specific domains in healthcare to validate the effectiveness of the proposed model. The ChatGPT application in case studies of healthcare is tested using datasets for English and Vietnamese languages. In comparative experimental testing, the proposed model outperformed the state of the art, achieving in the range of 96.70–97.50% performance using a heart dataset.

Variability in treatment effects in an English national dataset of psychological therapies: The relationships between severity, treatment duration, and therapy type

BackgroundComparisons between bona fide psychological therapies generally report no effects or small effects favoring cognitive behavioral therapy (CBT), suggesting that differences between therapy modalities are of little importance. MethodsAn observational, cohort study of patients (N = 11,116), drawn from the English Talking Therapies program. Patients presented with depression and were treated by CBT or person-centered experiential therapy (PCET), 67 % were female and the age range was 15–94. Multilevel modeling was used to identify variables associated with outcomes and CBT and PCET outcomes were compared dependent on the number of sessions patients attended and the severity of their depression at intake. ResultsAlthough overall a small effect size of 0.14 (0.10, 0.18) favored CBT, we found differing effect sizes depending on patient severity and the number of sessions they had attended at outcome. For non-clinical and moderately depressed patients no significant differences between therapies were found. For moderately-severe and severe patients there was a crossover in effectiveness with PCET more effective up to 5 or 6 sessions and CBT more effective thereafter. However, small effects in favour of PCET were unreliable. The largest effects were found for CBT for patients who were moderately-severe and had >10 sessions (d = 0.30, 0.12, 0.48) or were severe and had >12 sessions (d = 0.26 (0.02, 0.49). LimitationsNo data was available on therapy allocation decisions or therapists and the results may not be generalizable to non-NHS Talking Therapies services. ConclusionsSmall effects can mask important contextual differences between therapies and their study can inform allocation decisions.

Development of Pidgin English Hate Speech Classification System for Social Media

With the widespread use of social media, people from all walks of life—individuals, friends, family, public and private organizations, business communities, states, and entire nations—are exchanging information in various formats, including text, messages, audio, video, cartons, and pictures. Social media also facilitates the distribution and propagation of hate speech, despite the immense benefits of knowledge sharing through these platforms. The purpose of this work was to construct a text-based, Pidgin English hate speech classification system (HSCS) in social media, taking into account the alarming rate at which hate speech is shared and propagated on social media, as well as the negative effects of hate speech on society. We used text data sets in Pidgin English that were taken from Twitter and Facebook (3,153). To train the Support Vector Machine (SVM) text classifier to identify hate speech in Pidgin English, 70% of the Pidgin English data set was annotated. The SVM classifier's performance was tested and assessed using the remaining thirty percent of the Pidgin English text data set. The test set findings' confusion matrix, as determined by the HSCS performance evaluation, was 62.04%, 64.42%, 0.7541, 0.6947, and 0.64 in terms of accuracy, precision, recall, F1-score, and Receiver Operating Characteristics (ROC) curve. When HSCS was compared to other Machine Learning (ML) classifiers, such as Logistic Regression (LR), Random Forest (RF), and Naive Bayes, the results showed that LR had accuracy and precision of 61.51% and 63.89%, RF had 54.88% and 50.65%, and Naive Bayes had 61.51% and 63.89%.

Gender-Driven English Speech Emotion Recognition with Genetic Algorithm.

Speech emotion recognition based on gender holds great importance for achieving more accurate, personalized, and empathetic interactions in technology, healthcare, psychology, and social sciences. In this paper, we present a novel gender-emotion model. First, gender and emotion features were extracted from voice signals to lay the foundation for our recognition model. Second, a genetic algorithm (GA) processed high-dimensional features, and the Fisher score was used for evaluation. Third, features were ranked by their importance, and the GA was improved through novel crossover and mutation methods based on feature importance, to improve the recognition accuracy. Finally, the proposed algorithm was compared with state-of-the-art algorithms on four common English datasets using support vector machines (SVM), and it demonstrated superior performance in accuracy, precision, recall, F1-score, the number of selected features, and running time. The proposed algorithm faced challenges in distinguishing between neutral, sad, and fearful emotions, due to subtle vocal differences, overlapping pitch and tone variability, and similar prosodic features. Notably, the primary features for gender-based differentiation mainly involved mel frequency cepstral coefficients (MFCC) and log MFCC.

Location-enhanced syntactic knowledge for biomedical relation extraction

Biomedical relation extraction has long been considered a challenging task due to the specialization and complexity of biomedical texts. Syntactic knowledge has been widely employed in existing research to enhance relation extraction, providing guidance for the semantic understanding and text representation of models. However, the utilization of syntactic knowledge in most studies is not exhaustive, and there is often a lack of fine-grained noise reduction, leading to confusion in relation classification. In this paper, we propose an attention generator that comprehensively considers both syntactic dependency type information and syntactic position information to distinguish the importance of different dependency connections. Additionally, we integrate positional information, dependency type information, and word representations together to introduce location-enhanced syntactic knowledge for guiding our biomedical relation extraction. Experimental results on three widely used English benchmark datasets in the biomedical domain consistently outperform a range of baseline models, demonstrating that our approach not only makes full use of syntactic knowledge but also effectively reduces the impact of noisy words.

Quality achhi hai (is good), satisfied! Towards aspect based sentiment analysis in code-mixed language

Social media, e-commerce, and other online platforms have witnessed tremendous growth in multilingual users. This requires addressing the code-mixing phenomenon, i.e. mixing of more than one language for providing a rich native user experience. User reviews and comments may benefit service providers in terms of customer management. Aspect based Sentiment Analysis (ABSA) provides a fine-grained analysis of these reviews by identifying the aspects mentioned and classifies the polarities (i.e., positive, negative, neutral, and conflict). The research in this direction has mainly focused on resource-rich monolingual languages like English, which does not suffice for analyzing multilingual code-mixed reviews. In this paper, we introduce a new task to facilitate the research on code-mixed ABSA. We offer a benchmark setup by creating a code-mixed Hinglish (i.e., mixing of Hindi and English) dataset for ABSA, which is annotated with aspect terms and their sentiment values. To demonstrate the effective usage of the dataset, we develop several deep learning based models for aspect term extraction and sentiment analysis, and establish them as the baselines for further research in this direction. 11Codes and the complete dataset have been made available on https://www.iitp.ac.in/~ai-nlp-ml/resources.html and at Github repository: https://github.com/20118/ABSA-MIX.

Handwritten Signature Verification Method Using Convolutional Neural Network

Automatic signature verification methods play a significant role in providing a secure and authenticated handwritten signature in many applications, to prevent forgery problems, specifically institutions of finance, and transections of legal papers, etc. There are two types of handwritten signature verification methods: online verification (dynamic) and offline verification (static) methods. Besides, signature verification approaches can be categorized into two styles: writer dependent (WD), and writer independent (WI) styles. Offline signature verification methods demands a high representation features for the signature image. However, lots of studies have been proposed for WI offline signature verification. Yet, there is necessity to improve the overall accuracy measurements. Therefore, a proved solution in this paper is depended on deep learning via convolutional neural network (CNN) for signature verification and optimize the overall accuracy measurements. The introduced model is trained on English signature dataset. For model evaluation, the deployed model is utilized to make predictions on new data of Arabic signature dataset to classify whether the signature is real or forged. The overall obtained accuracy is 95.36\% based on validation dataset.

APTrans: Transformer-Based Multilayer Semantic and Locational Feature Integration for Efficient Text Classification

Text classification is not only a prerequisite for natural language processing work, such as sentiment analysis and natural language reasoning, but is also of great significance for screening massive amounts of information in daily life. However, the performance of classification algorithms is always affected due to the diversity of language expressions, inaccurate semantic information, colloquial information, and many other problems. We identify three clues in this study, namely, core relevance information, semantic location associations, and the mining characteristics of deep and shallow networks for different information, to cope with these challenges. Two key insights about the text are revealed based on these three clues: key information relationship and word group inline relationship. We propose a novel attention feature fusion network, Attention Pyramid Transformer (APTrans), which is capable of learning the core semantic and location information from sentences using the above-mentioned two key insights. Specially, a hierarchical feature fusion module, Feature Fusion Connection (FFCon), is proposed to merge the semantic features of higher layers with positional features of lower layers. Thereafter, a Transformer-based XLNet network is used as the backbone to initially extract the long dependencies from statements. Comprehensive experiments show that APTrans can achieve leading results on the THUCNews Chinese dataset, AG News, and TREC-QA English dataset, outperforming most excellent pre-trained models. Furthermore, extended experiments are carried out on a self-built Chinese dataset theme analysis of teachers’ classroom corpus. We also provide visualization work, further proving that APTrans has good potential in text classification work.

Testing the viability of ChatGPT as a companion in L2 writing accuracy assessment

This study explores the effectiveness of ChatGPT as a tool for evaluating linguistic accuracy in second language (L2) writing, situated within the complexity, accuracy, and fluency (CAF) framework. By using the Cambridge Learner Corpus First Certificate in English (CLC FCE) dataset, an error-tagged learner corpus, it compares ChatGPT's performance to human evaluators and Grammarly in assessing errors or accuracy rates across 232 writing samples. The findings indicate a strong correlation between ChatGPT's assessments and human accuracy ratings, demonstrating its precision in automated assessments. In comparison to Grammarly, ChatGPT shows a closer alignment with human judgments and students’ writing scores. Thus, ChatGPT can be a potential tool for enhancing efficiency in L2 research and L2 writing pedagogy.

Large Language Models and the Wisdom of Small Crowds.

Recent advances in Large Language Models (LLMs) have raised the question of replacing human subjects with LLM-generated data. While some believe that LLMs capture the "wisdom of the crowd"-due to their vast training data-empirical evidence for this hypothesis remains scarce. We present a novel methodological framework to test this: the "number needed to beat" (NNB), which measures how many humans are needed for a sample's quality to rival the quality achieved by GPT-4, a state-of-the-art LLM. In a series of pre-registered experiments, we collect novel human data and demonstrate the utility of this method for four psycholinguistic datasets for English. We find that NNB > 1 for each dataset, but also that NNB varies across tasks (and in some cases is quite small, e.g., 2). We also introduce two "centaur" methods for combining LLM and human data, which outperform both stand-alone LLMs and human samples. Finally, we analyze the trade-offs in data cost and quality for each approach. While clear limitations remain, we suggest that this framework could guide decision-making about whether and how to integrate LLM-generated data into the research pipeline.

Albanian Handwritten Text Recognition using Synthetic Datasets and Pre-Trained Models

Handwritten Text Recognition (HTR) has continuously attracted the focus of researchers to enable the integration of technology into our daily lives. Handwritten text recognition (HTR), a technology of considerable importance, takes a leading role in the analysis and digitization of various documents. This technology is important in facilitating the efficient use of handwritten documents, especially within academic, historical, and cultural contexts. The use of artificial intelligence in handwriting recognition offers a very good opportunity to achieve satisfactory results in this field, but to achieve good results a large dataset is needed. Creating a large dataset to train different AI models is a challenge for languages with limited resources such as the Albanian language. This paper aims to present a novel approach to the development of an HTR system for the Albanian language using an attention-based encoder-decoder architecture. The dataset used in the experiments is a synthetic dataset generated using deep learning techniques based on the English language dataset as they are both variants of the Latin alphabet. We enhanced the dataset with two letters specific to Albanian, (“ë” and “ç”). The usage of pre-trained English models for handwriting recognition improved our model’s performance. The results of the experiments are very promising and prove that our approach is efficient in recognizing handwriting in the Albanian language. This shows that the attention-based encoder-decoder architecture can be adopted for different languages with limited resources.

SC-Net: Multimodal metaphor detection using semantic conflicts

The phenomenon of metaphors is often regarded as a departure from semantic selection and common sense in semantic composition tasks, with their fundamental characteristics being semantic conflicts. Recognizing intricate and elusive metaphors has consistently posed a formidable challenge in the realm of natural language processing. In this paper, we introduce a novel multimodal metaphor detection model named SC-Net. This model benefits from linguistic metaphor identification theories and aims to exploit the inherent semantic conflicts in metaphors to detect whether multimodal corpora are metaphorical. The model projects each modality into two distinct subspaces: a modality-semantic feature space and a modality-latent metaphorical feature space. The features derived from these spaces provide a comprehensive perspective for capturing both intramodal semantic conflicts and intermodal semantic conflicts for prediction purposes. Our experiments conducted on the Chinese, English, and Bilingual Met-meme datasets and the MultiBully dataset demonstrate that our proposed SC-Net achieves state-of-the-art performance.

XTime: A general rule-based method for time expression recognition and normalization

Time expression (a.k.a., timex) recognition and normalization (TERN) is a crucial task for downstream research. However, previous studies have overlooked the critical characteristics of timexes that significantly impact the task. To gain deeper insights, we conduct an analysis across four diverse English datasets to examine the key attributes of timex constituents. Our analysis reveals several noteworthy observations, such as: timexes tend to very short; the majority of timexes contain time tokens; there exist strong mapping relationships between time tokens and timex types; there exists a priority relationship among timex types; and timex values exhibit only some standard formats. Based on these insights, we propose a novel general rule-based method termed XTime11Source codes and datasets are available at https://github.com/xszhong/XTime. to recognize timexes from free text and normalize them into standard formats. Notably, XTime’s rules are designed in a general and heuristic manner, enabling its independence of diverse domains and text types. Experimental evaluations conducted on both in-domain and out-of-domain English datasets demonstrate that XTime consistently outperforms or performs comparably to representative state-of-the-art methods.

More Than Syntaxes: Investigating Semantics to Zero-shot Cross-lingual Relation Extraction and Event Argument Role Labelling

Syntactic dependency structures are commonly utilized as language-agnostic features to solve the word order difference issues in zero-shot cross-lingual relation and event extraction tasks. However, while sentences in multiple forms can be employed to express the same meaning, the syntactic structure may vary considerably in specific scenarios. To fix this problem, we find semantics are rarely considered, which could provide a more consistent semantic analysis of sentences and be served as another bridge between different languages. Therefore, in this article, we introduce Syntax and Semantic Driven Network (SSDN) to equip syntax and semantic knowledge across languages simultaneously. Specifically, predicate–argument structures from semantic role labelling are explicitly incorporated into word representations. Then, a semantic-aware relational graph convolutional network and a transformer-based encoder are utilized to model both semantic dependency and syntactic dependency structures, respectively. Finally, a fusion module is introduced to integrate output representations adaptively. We conduct experiments on the widely used Automatic Content Extraction 2005 English, Chinese, and Arabic datasets. The evaluation results demonstrate that the proposed method achieves the state-of-the-art performance. Further study also indicates SSDN could produce robust representations that facilitate the transfer operations across languages.

MRMI-TTS: Multi-Reference Audios and Mutual Information Driven Zero-Shot Voice Cloning

Voice cloning in text-to-speech (TTS) is the process of replicating the voice of a target speaker with limited data. Among various voice cloning techniques, this article focuses on zero-shot voice cloning. Although existing TTS models can generate high-quality speech for seen speakers, cloning the voice of an unseen speaker remains a challenging task. The key aspect of zero-shot voice cloning is to obtain a speaker embedding from the target speaker. Previous works have used a speaker encoder to obtain a fixed-size speaker embedding from a single reference audio unsupervised, but they suffer from insufficient speaker information and content information leakage in speaker embedding. To address these issues, this article proposes MRMI-TTS, a FastSpeech2-based framework that uses speaker embedding as a conditioning variable to provide speaker information. The MRMI-TTS extracts speaker embedding and content embedding from multi-reference audios using a speaker encoder and a content encoder. To obtain sufficient speaker information, multi-reference audios are selected based on sentence similarity. The proposed model applies mutual information minimization on the two embeddings to remove entangled information within each embedding. Experiments on the public English dataset VCTK show that our method can improve synthesized speech in terms of both similarity and naturalness, even for unseen speakers. Compared to state-of-the-art reference embedding learned methods, our method achieves the best performance on the zero-shot voice cloning task. Furthermore, we demonstrate that the proposed method has a better capability of maintaining the speaker embedding in different languages. Sample outputs are available on the demo page. 1