Effective Model Research Articles

Non-Autoregressive Line-Level Code Completion

Software developers frequently use code completion tools to accelerate software development by suggesting the following code elements. Researchers usually employ AutoRegressive (AR) decoders to complete code sequences in a left-to-right, token-by-token fashion. To improve the accuracy and efficiency of code completion, we argue that tokens within a code statement have the potential to be predicted concurrently. In this article, we first conduct an empirical study to analyze the dependency among the target tokens in line-level code completion. The results suggest that it is potentially practical to generate all statement tokens in parallel. To this end, we introduce SANAR, a simple and effective syntax-aware non-autoregressive model for line-level code completion. To further improve the quality of the generated code, we propose an adaptive and syntax-aware sampling strategy to boost the model’s performance. The experimental results obtained from two widely used datasets indicate that our model outperforms state-of-the-art code completion approaches of similar model size by a considerable margin, and is faster than these models with up to 9× speed-up. Moreover, the extensive results additionally demonstrate that the enhancements achieved by SANAR become even more pronounced with larger model sizes, highlighting their significance.

Cross-sectoral role and community empowerment in controlling schistosomiasis in Indonesia

The role of cross-sectors in synergy with community empowerment in controlling schistosomiasis is a very effective model. Until mid-2017, the incidence rate of disease in humans in 28 endemic villages still ranged 0.00-2.15%. This research aims to evaluate schistosomiasis control activities that have been carried out across sectors based on a road map for schistosomiasis eradication and community empowerment in controlling schistosomiasis. The study was mixed method. Data were obtained by in-depth interview with key informants and evaluation of the results of cross-sectoral and community activities. The research was carried out in an endemic area for schistosomiasis on the Bada Plateau, Indonesia, from January to November 2019. In conclusion, the potential of cross-sector collaboration in controlling schistosomiasis has not been fully realized due to budget constraints. Community empowerment is carried out to eliminate snail foci, increase the scope of feces collection and change community behavior for the better in order to avoid infection and always try to clean up snail foci.

DAGCN: Dynamic and Adaptive Graph Convolutional Network for Salient Object Detection.

Deep-learning-based salient object detection (SOD) has achieved significant success in recent years. The SOD focuses on the context modeling of the scene information, and how to effectively model the context relationship in the scene is the key. However, it is difficult to build an effective context structure and model it. In this article, we propose a novel SOD method called dynamic and adaptive graph convolutional network (DAGCN) that is composed of two parts, adaptive neighborhood-wise graph convolutional network (AnwGCN) and spatially restricted K-nearest neighbors (SRKNN). The AnwGCN is novel adaptive neighborhood-wise graph convolution, which is used to model and analyze the saliency context. The SRKNN constructs the topological relationship of the saliency context by measuring the non-Euclidean spatial distance within a limited range. The proposed method constructs the context relationship as a topological graph by measuring the distance of the features in the non-Euclidean space, and conducts comparative modeling of context information through AnwGCN. The model has the ability to learn the metrics from features and can adapt to the hidden space distribution of the data. The description of the feature relationship is more accurate. Through the convolutional kernel adapted to the neighborhood, the model obtains the structure learning ability. Therefore, the graph convolution process can adapt to different graph data. Experimental results demonstrate that our solution achieves satisfactory performance on six widely used datasets and can also effectively detect camouflaged objects. Our code will be available at: https://github.com/CSIM-LUT/DAGCN.git.

Fault Detection of Urban Wastewater Treatment Process Based on Combination of Deep Information and Transformer Network.

As one of the hot issues of concerns during modern social development, the wastewater treatment process is acknowledged to be a process with complex biochemical reactions and susceptible to an external environment, featuring strong nonlinear and time correlation characteristics, which are difficult for traditional mechanism-based models to tackle. For many classical data-driven fault detection methods, a complete retraining process is necessary to monitor every new fault, and most of the current neural network-based strategies rarely achieve satisfactory monitoring accuracy or robustness either. Giving full consideration to the aforementioned problems, this article takes advantage of position encoding, residual connection, and multihead attention mechanism embedded in the Transformer structure to establish an effective and efficient wastewater treatment process fault detection model, where offline modeling and online monitoring are performed successively to achieve accurate detection of the faults. In the experimental part, the advantages of the proposed method are strongly verified through the simulation monitoring of 27 faults on the benchmark simulation model 1 (BSM1), where the false alarm rate (FAR) and miss alarm rate (MAR) of the established method are proved to be significantly lower than those of the compared state-of-the-art methods.

Uncertainty-Aware Aggregation for Federated Open Set Domain Adaptation.

Open set domain adaptation (OSDA) methods have been proposed to leverage the difference between the source and target domains, as well as to recognize the known and unknown classes in the target domain. Such methods typically require the entire source and target data simultaneously to train the target model. However, in real scenarios, data are distributed and stored in various clients. They cannot be exchanged among clients because of privacy protection. Federated learning (FL) is a decentralized approach for training an effective global model with the training data distributed among the clients. Despite its potential in addressing the privacy concerns of data sharing, FL methods for OSDA that can handle unknown classes is not yet available. To tackle this problem, we have developed a novel federated OSDA (FOSDA) algorithm. More specifically, FOSDA adopts an uncertainty-aware mechanism to generate a global model from all client models. It reduces the uncertainty of the federated aggregation by focusing on the contribution of source clients with high uncertainty while retaining those with high consistency. Moreover, a federated class-based weighted strategy is also implemented in FOSDA to maintain the category information of the source clients. We have conducted comprehensive experiments on three benchmark datasets to evaluate the performance of the proposed method, and the results demonstrate the effectiveness of FOSDA.

PDHL-MARCOS method based on cumulative prospect theory for MAGDM and its application in network security evaluation

The increasingly large and complex network brings convenience to our life and production, but it also brings potential dangers. Any minor vulnerability may lead to failure of application software configuration, software maintenance, user management and system management. Eventually, our network is interrupted, and even life and production data are leaked. Therefore, establishing a scientific and effective network security evaluation model is an effective way to supervise, manage and improve network security; It is also an effective measure to reduce the loss of life and production. Therefore, we first selected six factors closely related to network security to form a complete network security evaluation index system. Then, the utility function in the traditional measurement of alternatives and ranking according to compromise solution (MARCOS) method is improved by using the value function in cumulative prospect theory (CPT), and a PDHL-MARCOS evaluation model that can reflect the decision-makers’ (DMs’) attitudes towards gains and losses is obtained in the probabilistic double hierarchy linguistic term set (PDHLTS). In addition, the relative entropy (RE) method is also used to obtain expert weights. The combination weights of attributes are obtained using the correlation coefficient and the standard deviation (CCSD) method and the weight function in CPT.

CMCEE: A joint learning framework for cascade decoding with multi-feature fusion and conditional enhancement for overlapping event extraction

Event extraction (EE) is an important natural language processing task. With the passage of time, many powerful and effective models for event extraction tasks have been developed. However, there has been limited research on complex overlapping event extraction. Therefore, we propose a new cascade decoding model: A Joint Learning Framework for Cascade Decoding with Multi-Feature Fusion and Conditional Enhancement for Overlapping Event Extraction. 1) In this model, we introduce a cascade decoding mechanism with multi-feature fusion to better capture the interaction between decoding layers. 2) Additionally, we introduce an enhanced conditional layer normalization (ECLN) mechanism to enhance the interaction between subtasks. Simultaneously, the use of a cascade decoding model effectively addresses the problem of overlapping events. The model successively performs three subtasks, type detection, trigger word extraction and argument extraction. All three subtasks learned together in a framework, and a new conditional normalization mechanism is used to capture dependencies among these subtasks. The experiments are conducted using the overlapping event benchmark, FewFC dataset. The experimental evaluation demonstrates that our model achieves a higher F1 score on the overlapping event extraction task compared to the original overlapping event extraction model.

Silver-graphene oxide nanocomposite doping chitosan/PVA membrane for arsenic (III) elimination from aqueous solution

Heavy metals and pathogens from contaminated water sources may undoubtedly be removed by creating an efficient bio-adsorbent based on functional spots. Thus, the goal of this work was to produce chitosan (Ch)-polyvinyl alcohol (PVA) biofilm decorated with graphene oxide (GO) sheets doped with silver nanoparticles (AgNPs). The nanostructure of prepared GO/Ag nanosheets is examined by transmission electron microscope (TEM). The fabricated film (GO/Ag Ch-PVA) is compared by the control films (Ch, PVA and Ch-PVA). Fourier transform infrared (FTIR) spectroscopy, X-ray diffraction (XRD), and tensile strength are used to study the films’ structure. Also, the antimicrobial activity was assessed for the films. After doping the polymer matrix with GO/Ag, it was discovered that the tensile strength increased to about 46.18 MPa. Moreover, adsorption experiment for arsenic As (III) ions is explored by the prepared film at different operating conditions. The obtained results validated the enhanced adsorption ability of the GO/Ag Ch-PVA film towards As (III) with the highest adsorption capacity of 54.3 mg g−1 obtained from the isotherm model of Langmuir. Moreover, kinetic mathematical models for the adsorption effectiveness of GO/Ag Ch-PVA film are assessed. The results gathered demonstrated that GO/Ag Ch-PVA film is a potentially useful material for eliminating As (III) and microbial strains from essential water resources.

Fast Graph Generation via Spectral Diffusion.

Generating graph-structured data is a challenging problem, which requires learning the underlying distribution of graphs. Various models such as graph VAE, graph GANs, and graph diffusion models have been proposed to generate meaningful and reliable graphs, among which the diffusion models have achieved state-of-the-art performance. In this paper, we argue that running full-rank diffusion SDEs on the whole graph adjacency matrix space hinders diffusion models from learning graph topology generation, and hence significantly deteriorates the quality of generated graph data. To address this limitation, we propose an efficient yet effective Graph Spectral Diffusion Model (GSDM), which is driven by low-rank diffusion SDEs on the graph spectrum space. Our spectral diffusion model is further proven to enjoy a substantially stronger theoretical guarantee than standard diffusion models. Extensive experiments across various datasets demonstrate that our proposed GSDM turns out to be the SOTA model, by exhibiting both significantly higher generation quality and much less computational consumption than the baselines.

Collaborative Knowledge Distillation via Multiknowledge Transfer.

Knowledge distillation (KD), as an efficient and effective model compression technique, has received considerable attention in deep learning. The key to its success is about transferring knowledge from a large teacher network to a small student network. However, most existing KD methods consider only one type of knowledge learned from either instance features or relations via a specific distillation strategy, failing to explore the idea of transferring different types of knowledge with different distillation strategies. Moreover, the widely used offline distillation also suffers from a limited learning capacity due to the fixed large-to-small teacher-student architecture. In this article, we devise a collaborative KD via multiknowledge transfer (CKD-MKT) that prompts both self-learning and collaborative learning in a unified framework. Specifically, CKD-MKT utilizes a multiple knowledge transfer framework that assembles self and online distillation strategies to effectively: 1) fuse different kinds of knowledge, which allows multiple students to learn knowledge from both individual instances and instance relations, and 2) guide each other by learning from themselves using collaborative and self-learning. Experiments and ablation studies on six image datasets demonstrate that the proposed CKD-MKT significantly outperforms recent state-of-the-art methods for KD.

Data Augmentation for Sample Efficient and Robust Document Ranking

Contextual ranking models have delivered impressive performance improvements over classical models in the document ranking task. However, these highly over-parameterized models tend to be data-hungry and require large amounts of data even for fine-tuning. In this article, we propose data-augmentation methods for effective and robust ranking performance. One of the key benefits of using data augmentation is in achieving sample efficiency or learning effectively when we have only a small amount of training data. We propose supervised and unsupervised data augmentation schemes by creating training data using parts of the relevant documents in the query-document pairs. We then adapt a family of contrastive losses for the document ranking task that can exploit the augmented data to learn an effective ranking model. Our extensive experiments on subsets of the MS MARCO and TREC-DL test sets show that data augmentation, along with the ranking-adapted contrastive losses, results in performance improvements under most dataset sizes. Apart from sample efficiency, we conclusively show that data augmentation results in robust models when transferred to out-of-domain benchmarks. Our performance improvements in in-domain and more prominently in out-of-domain benchmarks show that augmentation regularizes the ranking model and improves its robustness and generalization capability.

Overcoming Barriers to Collaborative Team Practice: A System Approach.

All healthcare systems are challenged to provide sufficient access to appropriate care for the individuals and communities they serve. Among the commonly discussed interventions in an era of clinical shortages is the establishment of team-based care, where team members can practice at the top of their license. This solution ensures that talent and ability are amplified. However, this vision has been thwarted by several barriers. Recognizing that complex causes prevent team-based care, CommonSpirit Health leadership has embarked on a multifaceted implementation of tactics to mitigate the barriers. The strategy addresses issues ranging from regulations, laws, and payment practices to a lack of knowledge and understanding between professions. A combination of solutions, rather than discrete tactics, holds the key to the system's strategy. The complement of actions put in place at CommonSpirit Health has included new executive roles, dyad leadership models, shared multidisciplinary education, targeted advocacy, best practice playbooks, and the flexibility to individualize local models as part of the journey to a more effective and efficient model of care.

Improving student higher order thinking skills using SynecticHOTS-oriented learning model

The achievement of students’ learning outcomes in physics lessons has a low tendency, this is due to various factors, one of which is higher-order thinking skills (HOTS) based learning that has not been fully implemented in the learning process and assessment. This study aimed to determine the differences in the Synectic-HOTS learning model in physics compared to conventional learning models. This study used a quasi-experimental study with a pre-test–post-test nonequivalent control group design. The instrument used in this study is a two-tier test question that contains 20 items of HOTS aspect. The prerequisite test showed that the data was not homogeneous, while the results of the data homogeneity test were stated to be homogeneous. Therefore, a non-parametric test was carried out using the Mann-Whitney U-test. Based on the results of the data analysis, the Asymp. Sig. (2-tailed) for the pre-test was 0.111 and the post-test was 0.001 (p=.000<05). The effect size calculation shows that the Synectic learning model is effective in the medium category. These results indicate that the application of the SynecticHOTS model significantly improves HOTS when compared to conventional learning models. This work can open up insights to teachers about alternative, creative, and effective learning models to improve higher-order thinking skills and student learning outcomes.

CNN-CatBoost ensemble deep learning model for enhanced disease detection and classification of kidney disease

An efficient deep-learning prediction model for identifying chronic kidney disease (CKD) from exhaled breath is presented in this paper. The concentration of urea will be higher in CKD patients. Salivary urease breaks down the stored urea into ammonia, which is then excreted through breath. Thus, by monitoring the breath ammonia content, it is possible to identify the presence of high urea levels in the body. In this work, a novel sensing module is developed and applied to measure and assess the amount of ammonia in exhaled breath. Moreover, an effective deep learning prediction model that combines the CatBoost algorithm and convolutional neural network (CNN) is used to automate the prediction of disease. The proposed model, which combines the benefits of gradient-boosting and CNN, attained an exceptional accuracy of 98.37%. Experiments are conducted to evaluate the proposed model using real-time data and to assess how well it performs in comparison with existing deep learning methods. Our study's findings demonstrate that kidney disease can be accurately and noninvasively diagnosed using the proposed approach.

Hierarchical and dynamic graph attention network for drug-disease association prediction.

In the realm of biomedicine, the prediction of associations between drugs and diseases holds significant importance. Yet, conventional wet lab experiments often fall short of meeting the stringent demands for prediction accuracy and efficiency. Many prior studies have predominantly focused on drug and disease similarities to predict drug-disease associations, but overlooking the crucial interactions between drugs and diseases that are essential for enhancing prediction accuracy. Hence, in this paper, a resilient and effective model named Hierarchical and Dynamic Graph Attention Network (HDGAT) has been proposed to predict drug-disease associations. Firstly, it establishes a heterogeneous graph by leveraging the interplay of drug and disease similarities and associations. Subsequently, it harnesses the capabilities of graph convolutional networks and bidirectional long short-term memory networks (Bi-LSTM) to aggregate node-level information within the heterogeneous graph comprehensively. Furthermore, it incorporates a hierarchical attention mechanism between convolutional layers and a dynamic attention mechanism between nodes to learn embeddings for drugs and diseases. The hierarchical attention mechanism assigns varying weights to embeddings learned from different convolutional layers, and the dynamic attention mechanism efficiently prioritizes inter-node information by allocating each node with varying rankings of attention coefficients for neighbour nodes. Moreover, it employs residual connections to alleviate the over-smoothing issue in graph convolution operations. The latent drug-disease associations are quantified through the fusion of these embeddings ultimately. By conducting 5-fold cross-validation, HDGAT's performance surpasses the performance of existing state-of-the-art models across various evaluation metrics, which substantiates the exceptional efficacy of HDGAT in predicting drug-disease associations.

A simplified classification computational model of opinion mining using deep learning

Opinion and attempts to develop an automated system to determine people's viewpoints towards various units such as events, topics, products, services, organizations, individuals, and issues. Opinion analysis from the natural text can be regarded as a text and sequence classification problem which poses high feature space due to the involvement of dynamic information that needs to be addressed precisely. This paper introduces effective modelling of human opinion analysis from social media data subjected to complex and dynamic content. Firstly, a customized preprocessing operation based on natural language processing mechanisms as an effective data treatment process towards building quality-aware input data. On the other hand, a suitable deep learning technique, bidirectional long short term-memory (Bi-LSTM), is implemented for the opinion classification, followed by a data modelling process where truncating and padding is performed manually to achieve better data generalization in the training phase. The design and development of the model are carried on the MATLAB tool. The performance analysis has shown that the proposed system offers a significant advantage in terms of classification accuracy and less training time due to a reduction in the feature space by the data treatment operation.

The Dilemma of Quantum Neural Networks.

The core of quantum machine learning is to devise quantum models with good trainability and low generalization error bounds than their classical counterparts to ensure better reliability and interpretability. Recent studies confirmed that quantum neural networks (QNNs) have the ability to achieve this goal on specific datasets. In this regard, it is of great importance to understand whether these advantages are still preserved on real-world tasks. Through systematic numerical experiments, we empirically observe that current QNNs fail to provide any benefit over classical learning models. Concretely, our results deliver two key messages. First, QNNs suffer from the severely limited effective model capacity, which incurs poor generalization on real-world datasets. Second, the trainability of QNNs is insensitive to regularization techniques, which sharply contrasts with the classical scenario. These empirical results force us to rethink the role of current QNNs and to design novel protocols for solving real-world problems with quantum advantages.

Optimal Switching Sequence MPC for Four-Leg Two-Level Grid-Connected Converters

This paper presents a versatile and effective model predictive current control strategy based on optimal switching sequences for four–leg two–level grid–connected converters. The proposed control methodology, presents an excellent dynamic performance, good tracking of high-frequency reference signals, fixed switching frequency, and well-formed harmonic spectrum, enabling the proposed controller to govern shunt active filters and distributed generation applications based on four-leg converters. Experimental results demonstrate the effectiveness and high-quality performance obtained with the proposed strategy.

Implementation of a PAFV turbulence model for airfoil flow simulation on OpenFOAM

To further develop a more effective turbulence model and to improve the calculation accuracy of the flow around airfoil, a new PAFV turbulence model has here been constructed by using a deformation rate tensor and the grouping of an average fluctuation velocity. To evaluate the applicability of the PAFV turbulence model, the numerical calculations of flow around the airfoil have here been implemented, which was based on the OpenFOAM calculation platform. On the basis of grid independence research, the model was used to calculate the low-speed flow-around problem for the plano-convex airfoil NACA4412 and the symmetric airfoil NACA0012. It was also compared with the S-A (Spalart-Allmaras) and SST (Shear Stress Transport) k-ω turbulence models. Firstly, the maximum lift angle-of-attack case of the NACA4412 airfoil was calculated. Thereafter, numerical calculations were performed for the flow around the airfoil in the multi-angle-of-attack case of the NACA0012 airfoil. The results showed that the NACA4412 airfoil had an obviously separated vortex at the trailing edge of the airfoil at the maximum lift angle of attack. Also, there was a certain velocity loss downstream of the trailing edge, which was calculated by all three models. However, the results of the PAFV turbulence model were found to be better than those of the S-A and SST turbulence models. The three turbulence models showed comparable accuracies for the calculations of the surface pressure coefficients of the NACA0012 airfoil. However, the S-A and SST k-ω turbulence models were slightly better for the calculations of the mean velocity profiles of the NACA0012 airfoil. Also, the PAFV turbulence model was more accurate for the calculations of the lift and drag coefficients. In conclusion, the PAFV model can make effective predictions of the airfoil low-speed flow around the problem at hand, which in turn preliminarily verifies the applicability of this turbulence model for the low-speed flow around the airfoil problems.

Social work collective leadership in a large urban hospital during an era of unprecedented transformational change

This Department of Social Work Services (the Department) is one of the largest hospital social work departments in the country, with almost 600 staff, in the largest of eight hospitals and one medical school of a major health system. The Department’s senior directors (three current and one former) share its innovative and effective collective leadership model that was initially developed in response to a confluence of unprecedented complex events stemming from the Covid-19 pandemic. The experiences of these co-leaders, an evaluation of the model, future internal considerations as its applicability to other health care settings will be discussed.