Context Of Patterns Research Articles

An Articulatory Analysis of American English Rhotics in Children With and Without a History of Residual Speech Sound Disorder.

This study investigated articulatory patterns for American English /ɹ/ in children with and without a history of residual speech sound disorder (RSSD). It was hypothesized that children without RSSD would favor bunched tongue shapes, similar to American adults reported in previous literature. Based on clinical cueing practices, it was hypothesized that children with RSSD might produce retroflex tongue shape patterns at a higher relative rate. Finally, it was hypothesized that, among children who use a mixture of bunched and retroflex shapes, phonetic context would impact tongue shape as reported in the adult literature. These hypotheses were tested using ultrasound data from a stimulability task eliciting /ɹ/ in syllabic, postvocalic, and onset contexts. Participants were two groups of children/adolescents aged 9-15 years: 36 with RSSD who completed a study of ultrasound biofeedback treatment and 33 with no history of RSSD. Tongue shapes were qualitatively coded as bunched or retroflex using a flowchart from previous research. Children with no history of RSSD were found to use bunched-only tongue shape patterns at a rate higher than adults, but those who used a mixture of shapes for /ɹ/ followed the expected phonetic contextual patterning. Children with RSSD were found to use retroflex-only patterns at a substantially higher rate than adults, and those using a mixture of shapes did not exhibit the expected patterning by phonetic context. These findings suggest that clients receiving ultrasound biofeedback treatment for /ɹ/ may be most responsive to clinician cueing of retroflex shapes, at least early on. However, retroflex-only cueing may be a limiting and insufficient strategy, particularly in light of our finding of a lack of typical variation across phonetic contexts in children with remediated /ɹ/. Future research should more specifically track cueing strategies to better understand the relationship between clinician cues, tongue shapes, and generalization across a range of contexts. https://doi.org/10.23641/asha.26801050.

Synergizing human insight and machine learning: A dual-lens approach to uncovering healthcare research and innovation outcomes

Many healthcare organisations have extensive documentation detailing the processes behind their various research and innovation projects. Analysing this data can provide valuable insights into why some projects succeed without major issues, others encounter and overcome problems, and some ultimately fail. This study introduces an approach that combines narrative interviews and Natural Language Processing (NLP) to identify patterns associated with innovation project outcomes. We analysed 618 documents from 67 projects provided by ZonMw, a major Dutch healthcare research funder, and conducted 32 narrative interviews across seven cases of healthcare innovation projects. By using narrative interviews to inform and pre-train a text embedding model, we demonstrate the potential to create a proxy for human judgement, allowing for a more natural identification of contextual patterns in project documentation. The findings indicate that successful projects are more likely to adopt a proactive approach to role changes and uncertainty (due to ambiguous laws and regulations) and to allow flexibility, which enhances stakeholder engagement, compared to failed projects. However, while we were able to conduct descriptive analysis to gain these insights, significant interpretation is still required to fully understand the findings. Our study makes two primary contributions: first, it offers a new approach for future research on the factors that determine project success or failure, closely aligning with Structuration Theory. Additionally, it suggests potential efficiency improvements in theory development by enabling multiple pattern configurations within Grounded Theory. Second, it offers practical strategies for organisations to more effectively capture and use contextual information in their project documentation for future success.

Leveraging GPT-4 for identifying cancer phenotypes in electronic health records: a performance comparison between GPT-4, GPT-3.5-turbo, Flan-T5, Llama-3-8B, and spaCy's rule-based and machine learning-based methods.

Accurately identifying clinical phenotypes from Electronic Health Records (EHRs) provides additional insights into patients' health, especially when such information is unavailable in structured data. This study evaluates the application of OpenAI's Generative Pre-trained Transformer (GPT)-4 model to identify clinical phenotypes from EHR text in non-small cell lung cancer (NSCLC) patients. The goal was to identify disease stages, treatments and progression utilizing GPT-4, and compare its performance against GPT-3.5-turbo, Flan-T5-xl, Flan-T5-xxl, Llama-3-8B, and 2 rule-based and machine learning-based methods, namely, scispaCy and medspaCy. Phenotypes such as initial cancer stage, initial treatment, evidence of cancer recurrence, and affected organs during recurrence were identified from 13646 clinical notes for 63 NSCLC patients from Washington University in St. Louis, Missouri. The performance of the GPT-4 model is evaluated against GPT-3.5-turbo, Flan-T5-xxl, Flan-T5-xl, Llama-3-8B, medspaCy, and scispaCy by comparing precision, recall, and micro-F1 scores. GPT-4 achieved higher F1 score, precision, and recall compared to Flan-T5-xl, Flan-T5-xxl, Llama-3-8B, medspaCy, and scispaCy's models. GPT-3.5-turbo performed similarly to that of GPT-4. GPT, Flan-T5, and Llama models were not constrained by explicit rule requirements for contextual pattern recognition. spaCy models relied on predefined patterns, leading to their suboptimal performance. GPT-4 improves clinical phenotype identification due to its robust pre-training and remarkable pattern recognition capability on the embedded tokens. It demonstrates data-driven effectiveness even with limited context in the input. While rule-based models remain useful for some tasks, GPT models offer improved contextual understanding of the text, and robust clinical phenotype extraction.

Gender Patterns in Indonesian: A Corpus Study of Personal Pronoun References “Ia” and “Dia”

Indonesian is a language that does not have a grammatical gender feature, and this absence extends to the third-person singular pronouns. In English and many European languages, there are distinct second-person for addressing genders: male and female. The grammatical structure of Indonesian remains unchanged despite temporal shifts and the absence of gender-specific distinctions. The official Indonesian dictionary (Kamus Besar Bahasa Indonesia or KBBI) does not designate ia and dia with explicit gender distinctions, such as masculine and feminine. This research aims to discuss the masculine and feminine contextual patterns of the pronouns ia and dia within the Indonesian Corpus (Korpus Bahasa Indonesia or KOIN). The study used a dataset comprising 121.098 tokens selected from literature, national, and social categories. Identifying gender patterns in using the second-person pronouns gets the research focus on the data of 19.697 concordances of the word ia and 10.031 concordances of dia. The findings indicate that dia has a reasonably equal association with both genders, but the word ia prefers feminine references more. It clarifies the nuances of gender expression in Indonesian and explains how linguistic decisions communicate gender information.

Multi-Label Action Anticipation for Real-World Videos With Scene Understanding.

With human action anticipation becoming an essential tool for many practical applications, there has been an increasing trend in developing more accurate anticipation models in recent years. Most of the existing methods target standard action anticipation datasets, in which they could produce promising results by learning action-level contextual patterns. However, the over-simplified scenarios of standard datasets often do not hold in reality, which hinders them from being applied to real-world applications. To address this, we propose a scene-graph-based novel model SEAD that learns the action anticipation at the high semantic level rather than focusing on the action level. The proposed model is composed of two main modules, 1) the scene prediction module, which predicts future scene graphs using a grammar dictionary, and 2) the action anticipation module, which is responsible for predicting future actions with an LSTM network by taking as input the observed and predicted scene graphs. We evaluate our model on two real-world video datasets (Charades and Home Action Genome) as well as a standard action anticipation dataset (CAD-120) to verify its efficacy. The experimental results show that SEAD is able to outperform existing methods by large margins on the two real-world datasets and can also yield stable predictions on the standard dataset at the same time. In particular, our proposed model surpasses the state-of-the-art methods with mean average precision improvements consistently higher than 65% on the Charades dataset and an average improvement of 40.6% on the Home Action Genome dataset.

Analysis of Tonality of Text Using Machine Learning

Abstract: Tonality analysis using machine learning algorithms refers to the computational examination of written text with the objective of discerning and categorizing the emotional or subjective tone embedded within the content. In the realm of natural language processing (NLP), this analytical approach seeks to automate the process of understanding and classifying the sentiment expressed in textual data, facilitating a nuanced interpretation of language that extends beyond mere lexical meaning. At its core, the analysis of tonality involves the utilization of sophisticated machine learning models to distinguish and categorize the emotional underpinnings of text. This is particularly valuable in scenarios where vast amounts of unstructured textual data need to be processed, such as customer reviews, social media interactions, or news articles. The overarching goal is to automatically assign predefined tonal labels, such as positive, negative, or neutral, to segments of text, thereby enabling a more quantitative and systematic investigation of the subjective aspects of language. The process typically commences with the acquisition of raw textual data, which undergoes a series of preprocessing steps. These steps involve cleaning the text, breaking it into individual units (tokenization), and extracting relevant linguistic features. Feature extraction is a critical aspect, as it involves transforming the raw text into a format suitable for machine learning algorithms. This may encompass capturing word frequencies, utilizing n-grams to identify contextual patterns, and incorporating sentiment lexicon scores to gauge the emotions.

Architecture of Narrative Writing in Saudi Novels Literature and Criticism

Via a systematic and architectural narrative vision through which the narrative movement is trace, this research is about architecture of Saudi novels, denoting contextual patterns from the first threshold (title), symbols, signs, and icons, from the preface and chapters to the distribution of the black ink and areas, to the architecture of horizontal and vertical writing, and punctuation and pauses as well as the elements of advertising, presentation, and focalization. Saudi novels are classified into tense narrative architecture, empirical narrative architecture, and artistic narrative architecture. Keywords: novel, Saudi Arabia, narration, architecture, writing, focalization

Barriers and enablers for oral health promotion programs amongst primary healthcare stakeholders in Qatar – a qualitative investigation

BackgroundOral health of preschool children remains a concern globally. Primary healthcare providers are in a vital position to support preventive oral care programmes. This study explored current practices, perception and barriers of primary health care professionals towards oral health promotion program of children in Qatar.MethodsThe qualitative research used focus group discussions and interviewed a total of 108 participants that were audio recorded and transcribed verbatim. Four major themes emerged and were analysed to explore contextual patterns within the data.ResultsParticipants acknowledged the high prevalence of caries in children and identified the causes in the local context which included parental practices, poor dietary habits, impact of culture lack of oral health knowledge, limitations in the healthcare system, and negative role of the media. However, complex barriers were exposed, including lack of time and ownership, system coordination between organizations, and lack of policy.ConclusionHealth professionals and bureaucrats involved in decision-making held a positive attitude towards oral health prevention programs and were enthusiastic to initiate and support these programs.

Strategic Navigations of Identity in the Face of Bi-Erasure: Narratives of Bisexual College Student Identity Negotiation

Despite constituting the largest segment of the LGBTQ + community, limited research exists on the experiences of bisexual university students, and that which does often includes them among aggregated samples of lesbian, gay, bisexual, transgender, and queer people. Using data from a critical poststructural narrative inquiry study guided by a queer theoretical framework, I sought to understand the unique experiences of bisexual students relative to their sexual identities and how they negotiate their bisexuality on the college campus. Sixteen participants engaged in semi-structured interviews and a drawing activity. Thematic and dialogic/performance analyses resulted in one contextual pattern: pervasive binegativity. Thematic analysis further revealed four patterns characterizing identity negotiation strategies, enacted within this broader context: (a) settling for simplicity, (b) downplaying bisexuality, (c) subtly signaling sexuality, and (d) outness as advocacy. Implications for research, theory, and practice are offered.

Sustainability and impact of the implementation of a frailty checklist for the acute medical unit: experience from a tertiary public hospital in Singapore

BackgroundAccelerated population ageing is associated with an increasing prevalence of frailty. International guidelines call for systematic assessment and timely interventions for older persons requiring acute care. Checklists have been applied...

Referring Expression Comprehension Using Language Adaptive Inference

Different from universal object detection, referring expression comprehension (REC) aims to locate specific objects referred to by natural language expressions. The expression provides high-level concepts of relevant visual and contextual patterns, which vary significantly with different expressions and account for only a few of those encoded in the REC model. This leads us to a question: do we really need the entire network with a fixed structure for various referring expressions? Ideally, given an expression, only expression-relevant components of the REC model are required. These components should be small in number as each expression only contains very few visual and contextual clues. This paper explores the adaptation between expressions and REC models for dynamic inference. Concretely, we propose a neat yet efficient framework named Language Adaptive Dynamic Subnets (LADS), which can extract language-adaptive subnets from the REC model conditioned on the referring expressions. By using the compact subnet, the inference can be more economical and efficient. Extensive experiments on RefCOCO, RefCOCO+, RefCOCOg, and Referit show that the proposed method achieves faster inference speed and higher accuracy against state-of-the-art approaches.

Semi-Supervised Air Quality Forecasting via Self-Supervised Hierarchical Graph Neural Network

Predicting air quality in fine spatiotemporal granularity is of great importance for air pollution control and urban sustainability. However, existing studies are either focused on predicting station-wise future air quality, or inferring current air quality for unmonitored regions. How to accurately forecast future air quality for these unmonitored regions in a fine granularity remains an unexplored problem. In this paper, we propose the Self-Supervised Hierarchical Graph Neural Network (SSH-GNN), for fine-grained air quality forecasting in a semi-supervised way. Specifically, to augment spatially sparse air quality observations, SSH-GNN first approximates the city-wide air quality distribution based on historical readings and various urban contextual factors (e.g., weather conditions and traffic flows). Then, we propose a hierarchical recurrent graph neural network to make city-wide predictions, which encodes the spatial hierarchy of urban regions for long-range spatiotemporal correlation modeling. Moreover, by leveraging spatiotemporal self-supervision strategies, SSH-GNN exploits both universal topological and contextual patterns to further enhance the forecasting effectiveness. Extensive experiments on two real-world datasets show that SSH-GNN significantly outperforms the state-of-the-art algorithms.

MCWS-Transformers: Towards an Efficient Modeling of Protein Sequences via Multi Context-Window Based Scaled Self-Attention.

This paper advances the self-attention mechanism in the standard transformer network specific to the modeling of the protein sequences. We introduce a novel context-window based scaled self-attention mechanism for processing protein sequences that is based on the notion of (i) local context and (ii) large contextual pattern. Both notions are essential to building a good representation for protein sequences. The proposed context-window based scaled self-attention mechanism is further used to build the multi context-window based scaled (MCWS) transformer network for the protein function prediction task at the protein sub-sequence level. Overall, the proposed MCWS transformer network produced improved predictive performances, outperforming existing state-of-the-art approaches by substantial margins. With respect to the standard transformer network, the proposed network produced improvements in F1-score of +2.30% and +2.08% on the biological process (BP) and molecular function (MF) datasets, respectively. The corresponding improvements over the state-of-the-art ProtVecGen-Plus+ProtVecGen-Ensemble approach are +3.38% (BP) and +2.86% (MF). Equally important, robust performances were obtained across protein sequences of different lengths.

Deep-Distributed-Learning-Based POI Recommendation Under Mobile-Edge Networks

With the rapid development of edge intelligence in wireless communication networks, mobile-edge networks (MENs) have been broadly discussed in academia. Supported by considerable geographical data acquisition ability of mobile Internet of Things (IoT), the MENs can also provide spatial locations-based social service to users. Therefore, suggesting reasonable points-of-interest (POIs) to users is essential to improve user experience of MENs. As the simple user-location data is usually sparse and not informative, existing literature attempted to extend feature space from two perspectives: 1) contextual patterns and 2) semantic patterns. However, previous approaches mainly focused on internal features of users, yet ignoring latent external features among them. To address this challenge, in this article, a deep distributed-learning-based POI recommendation (Deep-PR) method is proposed for situations of MENs. In particular, hidden feature components from both local and global subspaces are deeply abstracted via representative learning schemes. Besides, propagation operations are embedded to iteratively reoptimize expressions of the feature space. The successive effect of the above two aspects contributes a lot to more fine-grained feature spaces, so that a recommendation accuracy can be ensured. Two types of experiments are also carried out on three real-world data sets to assess both efficiency and stability of the proposed Deep-PR. Compared with seven typical baselines with respect to four evaluation metrics, obtained results of the overall performance of the Deep-PR are excellent.

MHAR: a novel convolutional recurrent model for recognizing motion-based human activity

With sensors built into smartphones and wearable devices, Human Activity Recognition (HAR) makes it possible to identify regular activities. One of the challenges is to access the motion information from activity data and extract the features from this information. A general model trained without motion features may perform poorly while recognizing the contextual patterns in human activities. In this paper, we proposed the mHAR (motion HAR) model for recognizing static, gradual, and dynamic motion activity to address this challenge. We examine the impact of the proposed model against the general non-convolutional and convolutional recurrent networks in terms of accuracy rates and F1 score. As an experimental dataset, the publicly accessible WISDM was employed. Additionally, the findings of the experiments demonstrate that, in contrast to the unbalanced class problem, motion features play a significant role in activity recognition by achieving accuracy rates of 92.29% and an average F1 score of 83.33%.

Toward Data-Driven Digital Therapeutics Analytics: Literature Review and Research Directions

With the advent of digital therapeutics (DTx), the development of software as a medical device (SaMD) for mobile and wearable devices has gained significant attention in recent years. Existing DTx evaluations, such as randomized clinical trials, mostly focus on verifying the effectiveness of DTx products. To acquire a deeper understanding of DTx engagement and behavioral adherence, beyond efficacy, a large amount of contextual and interaction data from mobile and wearable devices during field deployment would be required for analysis. In this work, the overall flow of the data-driven DTx analytics is reviewed to help researchers and practitioners to explore DTx datasets, to investigate contextual patterns associated with DTx usage, and to establish the (causal) relationship between DTx engagement and behavioral adherence. This review of the key components of data-driven analytics provides novel research directions in the analysis of mobile sensor and interaction datasets, which helps to iteratively improve the receptivity of existing DTx.

Stock market uncertainty determination with news headlines: A digital twin approach

<abstract><p>We present a novel digital twin model that implements advanced artificial intelligence techniques to robustly link news and stock market uncertainty. On the basis of central results in financial economics, our model efficiently identifies, quantifies, and forecasts the uncertainty encapsulated in the news by mirroring the human mind's information processing mechanisms. After obtaining full statistical descriptions of the timeline and contextual patterns of the appearances of specific words, the applied data mining techniques lead to the definition of regions of homogeneous knowledge. The absence of a clear assignment of informative elements to specific knowledge regions is regarded as uncertainty, which is then measured and quantified using Shannon Entropy. As compared with standard models, the empirical analyses demonstrate the effectiveness of this approach in anticipating stock market uncertainty, thus showcasing a meaningful integration of natural language processing, artificial intelligence, and information theory to comprehend the perception of uncertainty encapsulated in the news by market agents and its subsequent impact on stock markets.</p></abstract>

New field observations on the Quaternary geology and vertebrate palaeontological occurrences in the Narsinghpur region of Narmada valley (central India)

Abstract Since the beginning of the nineteenth century, the Narmada River Valley has been well known for discoveries of abundant vertebrate fossil localities, a wide range of prehistoric archaeological assemblages and the first-known hominin fossil site in India at Hathnora. The fossil and archaeological remains are found in various sedimentary contexts: boulder conglomerate, pebble-rich sand, cross-bedded, coarse-to-fine sand and reddish brown clay in the Early to Late Pleistocene deposits. Our preliminary field investigations in the Narsinghpur region of the basin mark the presence of 26 individual localities within a stretch of 70 km of the Narmada River and adjoining tributaries. Each location was documented by GPS and the geological details were recorded while collecting the fossils. Past and current palaeontological research has revealed that a variety of fauna adapted to the sub-humid climate of both aquatic and terrestrial landscapes that thrived in this area during the Pleistocene period. One of our long-term goals is to understand the reasons for the high number of palaeontological occurrences compared with other parts of India and use such information to predict the locations of new occurrences, including hominin fossils. Various taphonomic processes in this region are responsible for the preservation and destruction of fossil assemblages in different geological formations: soil type, climatic conditions, rich calcium carbonate presence, encrustation, patination and modern anthropogenic activities (among others). Most of the fossil localities are found at an elevation range of 310–350 m above mean sea level (AMSL). The Narmada Basin was inhabited by prehistoric human populations and a variety of fauna, as evidenced by the ample lithic and fossil assemblages across the region. There are definitive contextual patterns of occurrences of the Quaternary vertebrate fossils and archaeological assemblages in the landscape, which require detailed investigations and mapping to understand the spatial distribution pattern and nature of associated sedimentary environments. In addition, this uniquely fossil-rich area requires proper protection and long-term preservation as it is heavily impacted by modern anthropogenic factors such as agriculture, sand mining and other activities.

Health inequity in pandemic anxiety about COVID-19 infection and socioeconomic consequences in Japan: A structural equation modeling approach

BackgroundHealth inequity in relation to COVID-19 infection and socioeconomic consequences is a major global concern. Mental health issues in vulnerable populations have received special attention in research and practice during the COVID-19 pandemic. However, there is limited evidence on the nature of the anxieties experienced as a result of COVID-19, and how such concerns vary across demographic groups. AimThis study examines anxiety among the working population of Japan (aged 18–59), in terms of both COVID-19 infection and socioeconomic consequences, using an internationally validated tool, the Pandemic Anxiety Scale (PAS). MethodsData were collected using an online survey (n = 2,764). The analyses included an exploratory factor analysis (EFA), a confirmatory factor analysis (CFA), and structural equation modeling (SEM), followed by validation of the Japanese version of the PAS. ResultsA two-factor latent variable model shows the multidimensionality of anxiety in regard to the COVID-19 pandemic and the disparity across population groups in predicting the two defined anxiety dimensions. Several path coefficients showed somewhat unexpected and/or unique results from Japan compared with previous European studies. Specifically, self-reported health status was not significantly related to disease anxiety, and those who were not in paid employment reported lower consequence anxiety. The SEM results showed a greater number of significant exogenous variables for consequence anxiety compared to disease anxiety, highlighting disparities in pandemic anxiety by socioeconomic status in regard to socioeconomic consequences of the pandemic. ConclusionIn contrast to existing European studies, evidence from the current study suggests contextual patterns of health inequity. Due to the prolonged socioeconomic consequences of the pandemic, multidisciplinary research on mental health issues and the quality of life remains an important research agenda in exploring socioeconomic measures in context, towards addressing inequity concerns.

Deep Effective k-mer representation learning for polyadenylation signal prediction via co-occurrence embedding

The polyadenylation process begins as the transcription of a gene terminates, and polyadenylation signal prediction by the genomic sequence is the key to understanding the mechanism of mRNA metabolism. Computational algorithms are widely proposed for polyadenylation signal prediction, but they mainly rely on sequences’ statistical and compositional characteristics. In this work, we conduct adaptively context-aware feature learning for polyadenylation signal prediction via a deep model and co-occurrence embedding. Specifically, we devise an activation function with dynamic parameters to control the scale of features channel-wisely for the positive and negative part value of neuron outputs, and then use the statistical information of k-mer occurrences as features of polyadenylation data, which is pre-trained by an unsupervised learning algorithm. Next, by combining multiscale convolutions with adaptive coefficients, long short-term memory networks jointly extract spatial and temporal contextual patterns. The patterns are fused to obtain contextual information by an identity skip connection and addition strategy. Experimental results demonstrate that our proposed model achieves more remarkable performance than state-of-the-art methods across polyadenylation signal benchmarks, and an ablation study shows the effectiveness of the model, and k-mer embedding is a good feature representation for polyadenylation signals.