Your search

Medical classification is affected by many factors, and the traditional medical classification is usually restricted by factors such as too long text, numerous categories and so on. In order to solve these problems, this paper uses word vector and word vector to mine the text deeply, considering the problem of scattered key features of medical text, introducing long-term and short-term memory network to effectively retain the features of historical information in long text sequence, and using the structure of CNN to extract local features of text, through attention mechanism to obtain key features, considering the problems of many diseases, by using hierarchical classification. To stratify the disease. Combined with the above ideas, a deep DLCF model suitable for long text and multi-classification is designed. This model has obvious advantages in CMDD and other datasets. Compared with the baseline models, this model is superior to the baseline model in accuracy, recall and other indicators.

Hydrothermal activities on ultraslow-spreading ridges exhibit diverse characteristics, long histories with multiple participants, and might form large-scale, high-grade sulfide deposits. The Duanqiao hydrothermal field (DHF) is located at the segment with the thickest oceanic crust and a large axial magma chamber on the Southwest Indian Ridge, providing unique perspective of sulfide metallogenesis on ultraslow-spreading ridges. Previous studies revealed that DHF sulfide exhibits distinct features of enrichment of ore-forming elements in comparison with those of hydrothermal fields on sediment-starved mid-ocean ridges. However, the genesis and processes responsible for such differences remain poorly constrained. In this study, mineralogical, geochemical and S and Pb isotopic analyses were performed on relict sulfide mound samples to characterize DHF formation. The samples show clear concentric mineral zonation from the interior to the exterior wall. Assemblages of chalcopyrite, sphalerite, and pyrite are distributed mainly in the interior wall, whereas pyrite and marcasite are distributed mainly in the exterior wall. The low Cu content and Pb isotopic composition of the sulfide indicate that the metals are derived mainly from basement basalts. The δ34S values exhibit positive values distributed over a reasonably narrow range (2.42‰–7.97‰), which suggests approximately 62.1%–88.5% of S with basaltic origin. Compared with most hydrothermal fields along the sediment starved mid-ocean ridges, the DHF sulfide shows particularly high contents of Pb (263–2630 ppm), As (234–726 ppm), Sb (7.32–44.3 ppm), and Ag (35.2 to >100 ppm). The δ34S values exhibit an increasing tendency from the sample exterior to the interior. We propose that these features probably reflect the existence of a subsurface zone refining process. Our results provide new insight into the sulfide formation process and contribute to understanding the metallogenic mechanism of hydrothermal sulfides on ultraslow-spreading ridges.

As an emergent tourism sector, driving tourism connects car use and touristic activities intimately. Following the notion of the ‘inhabited car’, this article explores how and why Chinese tourists inhabit a travelling car for drivers/passengers in the leisure automobility and driving tourism context. Through three different road trips and ‘mobile methods’, it was found that Chinese tourists inhabit the car in four ways: driving, gazing, listening, and communicating. Through this embodied habitation, the car is turned into a ‘touristic inhabitation’ space for protecting the tourists generating touristic emotions、social interactions, and tourism meanings. The study contributes to automobility and tourism literature and provides implications for driving tourism development in China.

Nowadays, the increasing number of medical diagnostic data and clinical data provide more complementary references for doctors to make diagnosis to patients. For example, with medical data, such as electrocardiography (ECG), machine learning algorithms can be used to identify and diagnose heart disease to reduce the workload of doctors. However, ECG data is always exposed to various kinds of noise and interference in reality, and medical diagnostics only based on one-dimensional ECG data is not trustable enough. By extracting new features from other types of medical data, we can implement enhanced recognition methods, called multimodal learning. Multimodal learning helps models to process data from a range of different sources, eliminate the requirement for training each single learning modality, and improve the robustness of models with the diversity of data. Growing number of articles in recent years have been devoted to investigating how to extract data from different sources and build accurate multimodal machine learning models, or deep learning models for medical diagnostics. This paper reviews and summarizes several recent papers that dealing with multimodal machine learning in disease detection, and identify topics for future research.

<jats:p>Causal machine learning is an approach that combines causal inference and machine learning to understand and utilize causal relationships in data. In current research and applications, traditional machine learning and deep learning models always focus on prediction and pattern recognition. In contrast, causal machine learning goes a step further by revealing causal relationships between different variables. We explore a novel concept called Double Machine Learning that embraces causal machine learning in this research. The core goal is to select independent variables from a gesture identification problem that are causally related to final gesture results. This selection allows us to classify and analyze gestures more efficiently, thereby improving models’ performance and interpretability. Compared to commonly used feature selection methods such as Variance Threshold, Select From Model, Principal Component Analysis, Least Absolute Shrinkage and Selection Operator, Artificial Neural Network, and TabNet, Double Machine Learning methods focus more on causal relationships between variables rather than correlations. Our research shows that variables selected using the Double Machine Learning method perform well under different classification models, with final results significantly better than those of traditional methods. This novel Double Machine Learning-based approach offers researchers a valuable perspective for feature selection and model construction. It enhances the model’s ability to uncover causal relationships within complex data. Variables with causal significance can be more informative than those with only correlative significance, thus improving overall prediction performance and reliability.</jats:p>

Ligand peptides that have high affinity for ion channels are critical for regulating ion flux across the plasma membrane. These peptides are now being considered as potential drug candidates for many diseases, such as cardiovascular disease and cancers. In this work, we developed Multi-Branch-CNN, a CNN method with multiple input branches for identifying three types of ion channel peptide binders (sodium, potassium, and calcium) from intra- and inter-feature types. As for its real-world applications, prediction models that are able to recognize novel sequences having high or low similarities to training sequences are required. To this end, we tested our models on two test sets: a general test set including sequences spanning different similarity levels to those of the training set, and a novel-test set consisting of only sequences that bear little resemblance to sequences from the training set. Our experiments showed that the Multi-Branch-CNN method performs better than thirteen traditional ML algorithms (TML13), yielding an improvement in accuracy of 3.2%, 1.2%, and 2.3% on the test sets as well as 8.8%, 14.3%, and 14.6% on the novel-test sets for sodium, potassium, and calcium ion channels, respectively. We confirmed the effectiveness of Multi-Branch-CNN by comparing it to the standard CNN method with one input branch (Single-Branch-CNN) and an ensemble method (TML13-Stack). The data sets, script files to reproduce the experiments, and the final predictive models are freely available at https://github.com/jieluyan/Multi-Branch-CNN.

Antimicrobial resistance has become a critical global health problem due to the abuse of conventional antibiotics and the rise of multi-drug-resistant microbes. Antimicrobial peptides (AMPs) are a group of natural peptides that show promise as next-generation antibiotics due to their low toxicity to the host, broad spectrum of biological activity, including antibacterial, antifungal, antiviral, and anti-parasitic activities, and great therapeutic potential, such as anticancer, anti-inflammatory, etc. Most importantly, AMPs kill bacteria by damaging cell membranes using multiple mechanisms of action rather than targeting a single molecule or pathway, making it difficult for bacterial drug resistance to develop. However, experimental approaches used to discover and design new AMPs are very expensive and time-consuming. In recent years, there has been considerable interest in using in silico methods, including traditional machine learning (ML) and deep learning (DL) approaches, to drug discovery. While there are a few papers summarizing computational AMP prediction methods, none of them focused on DL methods. In this review, we aim to survey the latest AMP prediction methods achieved by DL approaches. First, the biology background of AMP is introduced, then various feature encoding methods used to represent the features of peptide sequences are presented. We explain the most popular DL techniques and highlight the recent works based on them to classify AMPs and design novel peptide sequences. Finally, we discuss the limitations and challenges of AMP prediction.

Many sports events have been reopened since the recovery from the global Covid19 epidemic. As countries work toward the third sustainable development goal of good health and well-being, sports events have received increasing support and investments. This study focuses on the Xiamen Marathon, which had reached the platinum level of sustainable development and is a benchmark marathon race in China. To assess the factors that influence the brand image and behavior of consumers of the Xiamen Marathon, this dissertation references the conceptual models in previous studies and hypothesizes that experience, brand image, loyalty, satisfaction, motivation, organization, quality, and trust are the direct influencing factors of brand image and indirect determinants of word of mouth, sport, and behavioral intention. This study surveyed participants of the Xiamen Marathon in 2024 and collected data from a sample of 285 participants. All the respondents were local residents and over 90% were contestants. The largest proportions were 36-to-55-year-old, married, highly educated, and employed, earning monthly income between ¥5000 to ¥15000 and spending over ¥1000 on marathon equipment. IBM SPSS and Amos were used for the quantitative analysis based on structural equation modeling techniques, including reliability test, validity test, and hypothesis testing. The questionnaire quality was good, the scale data were suitable for factor analysis, and the questionnaire data fitted well. However, quality and trust were excluded due to low significance. The results show that experience, loyalty, satisfaction, motivation, and organization have significant and positive impacts on brand image, which in turn has significant and positive impacts word of mouth, sport, and behavioral intention. The results support the eight hypotheses of direct effects and the 15 hypotheses of intermediate effects

In the age of the information society, the prominence of internet communication and social media has meant that the role of face-to-face encounters in public life has been diminished. This trend has been exacerbated by public health concerns in the aftermath of the COVID pandemic. The social role of architecture and its contribution to the sustainability of interpersonal relationships has become a significant issue for architects. As a place to accommodate sporting activities and social space in the community, sports centers play an important role in promoting social cohesion and interaction. This is the focal topic of the present thesis. The principal tasks of this thesis are as follows: (1) This thesis collects and studies relevant literature on sports centers as social spaces. A special emphasis is placed on discussions of social spaces. The intention is to articulate design characteristics of community sports centers that merit further exploration. (2) This thesis selects three representative sports facilities as case studies and considers how they promote the formation of social bonds and interactions. The thesis offers a review of design ideas, planning strategies and specific methods of implementation. (3) This thesis takes the design project for a sports center in New District A in Macau as an example, and puts forward a design strategy of ""vertical intensification + shared application"". This strategy aims to promote the integration of the sports center with the urban environment, while creating a flexible public space that can adapt to the needs of spontaneous and social activities. The intention is to make the sports center a place that promotes community interaction and connectivity. Overall, this thesis argues that community sports centers can become important places to promote social interaction through well-designed architectural space. Specific design factors and strategies explored in this thesis provide useful points of reference for architects. Through on-going processes of optimization and innovation, sports centers can better serve the community of Macau

With the comprehensive opening of the global economy, the intensification of market competition and the increasingly prominent value of human resources, all trades and professions are facing unprecedented challenges. In order to remain competitive in the market, human resource management plays an increasingly important role. As an important part of human resource management, enterprise performance management can be used as a key tool to drive the release of employees' potential, which has attracted more and more attention from enterprises. More and more performance management-related systems have been introduced into enterprises. However, in the system introduction and implementation process, there are different degrees of problems, such as ""mechanically copied"" and ""process"". Therefore, according to the specific situation of different enterprises, it is necessary to formulate the corresponding performance management system to maximize the role of motivating employees. This paper explores enterprise performance management's influence on the employee incentive effect. The implementation of enterprise performance management and its influence on employee incentive are analyzed through the literature review, case analysis and empirical analysis. Research has found that effective enterprise performance management can significantly improve the work enthusiasm and satisfaction of employees, thus enhancing the overall performance of enterprises and promoting all-round development. At the same time, this paper also discusses the key factors affecting the incentive effect of enterprise performance management, and puts forward the corresponding optimization suggestions

As environmental awareness grows, sustainable modes of transportation have garnered increasing importance. With the rising popularity of electric vehicles (EVs), a broader community now acknowledges their numerous advantages, such as lower noise levels, enhanced efficiency, and cost-effectiveness in comparison to traditional internal combustion engine vehicles. Furthermore, EVs significantly contribute to a more sustainable future by emitting fewer pollutants and reducing overall environmental impact. However, consumers' perceptions and expectations about the vehicle's country of origin - the nation where the vehicle is manufactured - remain unclear, particularly in the case of Chinese-manufactured EVs due to China's reputation for pollution and product safety concerns. This study employsa qualitative framework based on the Unified Theory of Acceptance and Use of Technology 2(UTAUT-2) to assess the acceptance of Chinese electric vehicles, and the influence of the Country-of-Origin Effect (COE) on a sample of Portuguese residents. Data was gathered through semi-structured interviews and analyzed using qualitative methods. The study's results shed light on the significance of the country of origin in shaping consumers' behavioral intentions to purchase Chinese electric vehicles, indicating a positive influence. This suggests that country of origin is a crucial factor when considering an individual's intention to adopt electric vehicles. Additionally, the research highlights the importance of various other factors such as performance expectancy, effort expectancy, social influence, facilitating conditions, price value, hedonic motivation, and habit in shaping consumers' attitudes and intentions. Our results underscore the complexity of consumer behavior toward electric vehicles, suggesting the need for a multifaceted approach to understanding andpromoting EV adoption. Research is needed to examine the COE in different cultural and geographic contexts to develop effective strategies to enhance the global adoption of electric vehicles, particularly from countries with varying environmental reputations

This study identifies Portuguese residents’ behavioural intention towards Chinese electric vehicles from the perspective of the Country-of-Origin Effect (COE) and uses the Unified Theory of Acceptance and Use of Technology 2 (UTAUT-2) as a guide to design interview questions. According to the existing literature, UTAUT-2, a comprehensive technology acceptance model, has not been used to assess the Portuguese electric vehicle market, and no relevant research has been found to apply the method of combining the COE into the UTAUT-2 framework. The purpose of this study is exploratory in nature, it uses qualitative methods to identify the behavioural intention of Portuguese residents towards the acceptance of Chinese electric vehicles. Data from 16 Portuguese residents was collected through semi-structured interviews and analyzed with qualitative methodology. The study found that factors such as performance expectations, environmental concerns, effort expectations, hedonic motivation, and social influence have a positive impact on Portuguese residents' purchase of electric vehicles, while price value, habits, and convenience conditions have a neutral or negative impact. Regarding COE, apart from social influence, it has no impact on other factors. To increase the popularity and sales of Chinese electric vehicles in Portugal, it is recommended that advertising and marketing efforts focus on price, charging services, after-sales service, and design