Node Weight Research Articles

A Hybrid Recognition Framework for Highly Interacting Machining Features Based on Primitive Decomposition, Learning and Reconstruction

For the highly interacting machining features, Layered Projection Decomposition Method presents inferior recognition efficiency and accuracy, due to its high-cost 3D projection and failures in determining projection faces for internal occluded faces. To address these issues, we propose a potential hybrid recognition framework. We first introduce a straightforward adjacent projection wire (APW) over UV wires, automatically restoring the full projection wires from highly interacting features. Building on APWs, an efficient hybrid boundary representation and its corresponding unambiguous primitive definitions are proposed by combining with graph-based boundary representations. Subsequently, we design an efficient primitive decomposition method by introducing primitive boundary matching to decide the initial projection faces, and introducing iterative projection boundary expansion to complete the full primitives from occluded faces. Moreover, we establish an efficient Graph Neural Network to learn the distinguishable distributions over the decomposed primitives. Specifically, an Adjacency Attention Unit is proposed to automatically perceive the influence weight of adjacent nodes, leading to more discriminative self-adaptive shape embedding for efficient primitive recognition. Finally, we summarize convenient reconstruction rules to correct the wrong predictions of feature faces with indistinguishable adjacent relationships. To evaluate the effectiveness of the proposed recognition framework, CAD models of complex aircraft structural parts are collected to present a challenging machining feature dataset. Extensive numerical experiments demonstrate that the proposed hybrid recognition framework enables significant improvements over the state-of-the-art machining feature recognition techniques.

Caloric Vestibular Stimulation Alleviates Cold Water Swimming Stress Induced Changes in BDNF and Haematological Parameters

Introduction: Stress is inevitable in daily life. Stress in long term has been implicated in the development of physical and neuronal disorders. Various studies have described the levels and role of BDNF in the brain following stress. Studies representing the changes in serum BDNF and wet weight of organs in stress are limited and controversial. Aim: The aim of the present study is to find the effect of stress on serum BDNF concentration, wet weight of selected organs, haematological parameters and to evaluate the effectiveness of caloric vestibular stimulation on stress-induced changes in these parameters. Methods and material: Male Wistar rats were exposed to cold water swimming stress for 14 days for a duration of 30 minutes per day. Serum corticosterone and serum BDNF were measured using the ELISA method. Weight of liver, spleen, thymus and lymph nodes were measured and their relation to body weight was calculated. Total RBC count, total WBC count, platelet count and haemoglobin concentration were analysed. Results: A statistically significant increase in serum corticosterone and BDNF was observed following stress. Liver weight, thymus weight and lymph nodes weight did not show any significant changes but spleen weight has reduced in the normal recovery group following stress. Total RBC count, total WBC count, platelet count and haemoglobin levels were significantly higher in the stress group. Caloric vestibular stimulation was effective in managing stress-induced changes in serum corticosterone, BDNF and spleen weight. Conclusion: Caloric vestibular stimulation is effective in restoring the stress-induced changes.

Component-based modeling of cascading failure propagation in directed dual-weight software networks

Software vulnerabilities often lead to cascading failures, resulting in service unavailability and potential breaches of user data. However, existing models for cascading failure propagation typically focus solely on the static design’s calling relationships, disregarding dynamic runtime propagation paths. Moreover, current network topology models primarily consider function calling frequency while overlooking critical factors like internal failure probability and component failure tolerance rates. Yet, these factors significantly influence the actual propagation of software cascading failures. In this study, we address these limitations by incorporating internal failure probabilities and calling frequencies as node and edge weights, respectively. This forms the basis of our component-based directed dual-weight software network cascading failure propagation model. This model encompasses the evaluation of cascading failure propagation through intra-component and inter-component propagation probabilities, alongside the constraint of component failure tolerance rates. Through extensive experiments conducted on six real-world software applications, our model has demonstrated its effectiveness in predicting software cascading failure propagation processes. This method deepens our understanding of software failures and structures, equipping software testers with the knowledge to make well-informed judgments regarding software quality concerns.

In Vivo Biocompatibility of Synechococcus sp. PCC 7002-Integrated Scaffolds for Skin Regeneration.

Cyanobacteria, commonly known as blue-green algae, are prevalent in freshwater systems and have gained interest for their potential in medical applications, particularly in skin regeneration. Among these, Synechococcus sp. strain PCC 7002 stands out because of its rapid proliferation and capacity to be genetically modified to produce growth factors. This study investigates the safety of Synechococcus sp. PCC 7002 when used in scaffolds for skin regeneration, focusing on systemic inflammatory responses in a murine model. We evaluated the following three groups: scaffolds colonized with genetically engineered bacteria producing hyaluronic acid, scaffolds with wild-type bacteria, and control scaffolds without bacteria. After seven days, we assessed systemic inflammation by measuring changes in cytokine profiles and lymphatic organ sizes. The results showed no significant differences in spleen, thymus, and lymph node weights, indicating a lack of overt systemic toxicity. Blood cytokine analysis revealed elevated levels of IL-6 and IL-1β in scaffolds with bacteria, suggesting a systemic inflammatory response, while TNF-α levels remained unaffected. Proteome profiling identified distinct cytokine patterns associated with bacterial colonization, including elevated inflammatory proteins and products, indicative of acute inflammation. Conversely, control scaffolds exhibited protein profiles suggestive of a rejection response, characterized by increased levels of cytokines involved in T and B cell activation. Our findings suggest that Synechococcus sp. PCC 7002 does not appear to cause significant systemic toxicity, supporting its potential use in biomedical applications. Further research is necessary to explore the long-term effects and clinical implications of these responses.

A simulation data-driven semi-supervised framework based on MK-KNN graph and ESSGAT for bearing fault diagnosis

Current supervised intelligent fault diagnosis relies on abundant labeled data. However, collecting and labeling data are typically both expensive and time-consuming. Fault diagnosis with unlabeled data remains a significant challenge. To address this issue, a simulation data-driven semi-supervised framework based on multi-kernel K-nearest neighbor (MK-KNN) and edge self-supervised graph attention network (ESSGAT) is proposed. The novel MK-KNN establishes the neighborhood relationships between simulation data and real data. The developed multi-kernel function mitigates the risks of overfitting and underfitting, thereby enhancing the robustness of the simulation-real graphs. The designed ESSGAT employs two forms of self-supervised attention to predict the presence of edges, increasing the weights of crucial neighboring nodes in the MK-KNN graph. The performance of the proposed method is evaluated using a public bearing dataset and a self-constructed dataset of high-speed train axle box bearings. The results show that the proposed method achieves better diagnostic performance compared with other state-of-the-art graph construction methods and graph convolutional networks.

DRADTiP: Drug repurposing for aging disease through drug-target interaction prediction

MotivationThe greatest risk factor for many non-communicable diseases is aging. Studies on model organisms have demonstrated that genetic and chemical perturbation alterations can lengthen longevity and overall health. However, finding longevity-enhancing medications and their related targets is difficult. MethodIn this work, we designed a novel drug repurposing model by identifying the interaction between aging-related genes or targets and drugs similar to aging disease. Each disease is associated with certain specific genetic factors for the occurrence of that disease. The factors include gene expression, pathway, miRNA, and degree of genes in the protein-protein interaction network. In this paper, we aim to find the drugs that prolong the life span of humans with their aging-related targets using the above-mentioned factors. In addition, the contribution or importance of each factor may vary among drugs and targets. Therefore, we designed a novel multi-layer random walk-based network representation learning model including node and edge weight to learn the features of drugs and targets respectively. ResultThe performance of the proposed model is demonstrated using k-fold cross-validation (k = 5). This model achieved better performance with scores of 0.93 and 0.91 for precision and recall respectively. The drugs identified by the system are evaluated to be potential candidates for aging since the degree of interaction between the potential drugs and their gene sets are high. In addition, the genes that are interacting with drugs produce the same biological functions. Hence the life span of the human will be increased or prolonged.

Weighted graph convolutional network with feature mask for low back pain prediction

Low back pain (LBP) is one of the most common physical disabilities worldwide, imposing a heavy medical economic burden. To help medical experts make disease predictions at an early stage, this paper proposes an LBP abnormality diagnosis model based on a weighted graph convolutional network with feature mask. Unlike popular neural network architectures, we represent the biomechanical characteristics of the spine as a weighted feature graph. The potential relationship between patients is mined through edge connections, and the importance of features for different patients is dynamically captured through the internal feature weighting of nodes. Finally, the importance score of each feature is calculated according to its identification and independence to generate the feature importance mask. It realizes feature sparsification and reinforces important features at the same time, so as to improve the accuracy and stability of the model. The proposed model is verified on a public dataset available in the Kaggle repository, and the classification accuracy achieves 90.2%, which is 2.14% higher than the current best result. Furthermore, when compared for stability against several commonly used machine learning algorithms, the standard deviation of accuracy over ten experiments is only about 1/3 of that of other algorithms.

Dynamic Hybrid Models With Active Sampling and Adaptive Selection of Double-Domain Features for the Tuning of Microwave Cavity Filters.

Microwave cavity filters are essential electromechanical coupling devices in communication systems. Structural-parameter tuning by experienced operators improves the filter performance but is demanding and time-consuming. The automatic tuning method has received extensive research attentions using data-driven modeling approaches. However, two main issues affect the accuracy and efficiency of the model construction: 1) features of tuning processes, as model inputs, have limited adaptability and extraction accuracy to different resonant states and 2) models require plentiful training data and the training process is time-consuming. Thus, dynamic hybrid models are developed in this study with self-selected inputs, self-organized samples, and a self-learning structure. First, spatial features are extracted to flexibly depict the tuning characteristic, and double-domain (spatial or circuital) features are selected adaptively to accommodate distinct resonance states. Second, a trustworthiness-curiosity-driven active sampling method is exploited to attain fewer and better-training data. Third, an improved glsms broad learning system acrlong BLS is developed using new modules of incremental node calculation and weight pruning, characterized by more lightweight and flexible structures. The proposed method is effective and flexible demonstrated by simulations and experiments, and the tuning task of microwave cavity filters is fulfilled in a more accurate and efficient manner.

Semantic Code Clone Detection Based on Community Detection

Semantic code clone detection is to find code snippets that are structurally or syntactically different, but semantically identical. It plays an important role in software reuse, code compression. Many existing studies have achieved good performance in non-semantic clone, but semantic clone is still a challenging task. Recently, several works have used tree or graph, such as Abstract Syntax Tree (AST), Control Flow Graph (CFG) or Program Dependency Graph (PDG) to extract semantic information from source codes. In order to reduce the complexity of tree and graph, some studies transform them into node sequences. However, this transformation will lose some semantic information. To address this issue, we propose a novel high-performance method that utilizes community detection to extract features of AST while preserving its semantic information. First, based on the AST of source code, we exploit community detection to split AST into different subtrees to extract the underlying semantics information of different code blocks, and use centrality analysis to quantify the semantic information as the weight of AST nodes. Then, the AST is converted into a sequence of tokens with weights, and a Siamese neural network model is used to detect the similarity of token sequences for semantic code clone detection. Finally, to evaluate our approach, we conduct experiments on two standard benchmark datasets, Google Code Jam (GCJ) and BigCloneBench (BCB). Experimental results show that our model outperforms the eight publicly available state-of-the-art methods in detecting code clones. It is five times faster than the tree-based method (ASTNN) in terms of time complexity.

A Recurrent Stochastic Configuration Network for Dynamic System Modeling and Its Application

In order to model nonlinear dynamic systems quickly and accurately, a recurrent stochastic configuration network (RSCN) with universal approximation ability is proposed in this paper. The method consists of network parameter learning and network structure determination. The network parameter learning strategy consists of input weights and the biases of hidden nodes generated randomly through a supervision mechanism, while the output weights of hidden nodes are determined by solving the least-squares problem. The network structure is determined incrementally by the training error of a single-layer recurrent neural network. The effectiveness of RSCN is tested and evaluated by using three nonlinear dynamic functions and historical data from a municipal solid waste incineration (MSWI) process. The experimental results show that the proposed method can improve modeling accuracy and reduce training time, which is valuable for applications in the field of industrial process modeling.

Network Topology Reconfiguration-Based Blind Equalization over Sensor Network.

Distributed in-network processing has garnered much attention due to its capability to estimate the unknown parameter of interest from noisy measurements based on a set of cooperating sensor nodes. In previous studies, distributed in-network processing mainly focused on short-distance communication systems, wherein sensor nodes collect certain parameters of interest within their maximum communication distance. In addition, the estimation of certain parameter vectors of interest from noisy measurements, relying heavily on training signals, is achieved with a non-blind distributed estimation algorithm. However, in some applications, acquiring knowledge of training signals beforehand is difficult. Therefore, it is necessary to perform distributed estimation algorithms for receivers without training signals, a concept known as blind distributed estimation. In this paper, the generalized Sato algorithm is used to design the blind equalizer for the signal estimation. In addition, we consider extending the short-distance communication system to a long-distance communication system for an unmanned aerial vehicle (UAV) cooperating with sensor nodes in the wireless sensor network (WSN). In this scenario, the data signal is transmitted from a UAV to the WSN and is received by sensor nodes. However, the performance of the blind equalizer is susceptible to the transmission channel in long-distance communication systems. Here, we present a network topology reconfiguration approach to address the issue of distributed blind equalization. The objective of the proposed method is to discard the influence of ill-channels on the other sensor nodes by detecting ill-channels and redesigning the sensor node weights. Through computer simulation experiments, we evaluated the performance of the blind equalizer using the average mean square error (MSE) and average symbol error rate (SER). In the results of the computer simulation experiments, the blind equalizer using the proposed method outperformed the conventional methods in terms of prediction accuracy and convergence speed.

Research on sales strategies for target customers of electric vehicles based on gradient boosting ensemble trees

This paper focuses on the automotive industry, especially the new energy vehicle sector, which is vital for the economy and needs careful market sales decisions. It analyzes data from 1,964 potential customers, including their satisfaction and personal information. The study uses machine learning to create a math model and give sales advice. The data is first cleaned without changing its original structure, and then normalized using the Z-Score method. For any outlier values in the data, the mean of similar items is used to fill in gaps. Specifically, for 500 missing entries about the number of children (B7), the paper groups customers by age, calculates the frequency of children in each group, and uses these frequencies to fill in missing values.Using the cleaned data, the XGBoost algorithm ranks the factors influencing customer decisions. For Brand 1, the top factors are economy (a3), work experience (B10), and safety performance (a4). For Brand 2, comfort (a2), disposable income (B15), and mortgage expenses (B16) are key. For Brand 3, car loan expenses (B17), battery technology (a1), and again mortgage expenses (B16) are most important.The paper also addresses imbalanced data by using an XGBoost binary classification model. This model predicts the likelihood of buying an electric vehicle using the Sigmoid function on tree node weights. It evaluates the model using Precision, Recall, and Area Under the Receiver Operating Characteristic Curve (AUC), which works well even with imbalanced data. The Precision and Recall rates for all three brands are above 90%, with AUC over 85% for Brands 1 and 2, and over 75% for Brand 3. This shows that the model is effective and can be used in real-world situations.

Gradient-free training of recurrent neural networks using random perturbations.

Recurrent neural networks (RNNs) hold immense potential for computations due to their Turing completeness and sequential processing capabilities, yet existing methods for their training encounter efficiency challenges. Backpropagation through time (BPTT), the prevailing method, extends the backpropagation (BP) algorithm by unrolling the RNN over time. However, this approach suffers from significant drawbacks, including the need to interleave forward and backward phases and store exact gradient information. Furthermore, BPTT has been shown to struggle to propagate gradient information for long sequences, leading to vanishing gradients. An alternative strategy to using gradient-based methods like BPTT involves stochastically approximating gradients through perturbation-based methods. This learning approach is exceptionally simple, necessitating only forward passes in the network and a global reinforcement signal as feedback. Despite its simplicity, the random nature of its updates typically leads to inefficient optimization, limiting its effectiveness in training neural networks. In this study, we present a new approach to perturbation-based learning in RNNs whose performance is competitive with BPTT, while maintaining the inherent advantages over gradient-based learning. To this end, we extend the recently introduced activity-based node perturbation (ANP) method to operate in the time domain, leading to more efficient learning and generalization. We subsequently conduct a range of experiments to validate our approach. Our results show similar performance, convergence time and scalability when compared to BPTT, strongly outperforming standard node perturbation and weight perturbation methods. These findings suggest that perturbation-based learning methods offer a versatile alternative to gradient-based methods for training RNNs which can be ideally suited for neuromorphic computing applications.

Software Fault Localization Based on Weighted Association Rule Mining and Complex Networks

Software fault localization technology aims to identify suspicious statements that cause software failures, which is crucial for ensuring software quality. Spectrum-based software fault location (SBFL) technology calculates the suspiciousness of each statement by analyzing the correlation between statement coverage information and execution results in test cases. SBFL has attracted increasing attention from scholars due to its high efficiency and scalability. However, existing SBFL studies have shown that a large number of statements share the same suspiciousness, which hinders software debuggers from quickly identifying the location of faulty statements. To address this challenge, we propose an SBFL model based on weighted association rule mining and complex networks: FL-WARMCN. The algorithm first uses Jaccard to measure the distance between passing and failing test cases, and applies it as the weight of passing test cases. Next, FL-WARMCN calculates the initial suspiciousness of each statement based on the program spectrum data. Then, the FL-WARMCN model utilizes a weighted association rule mining algorithm to obtain the correlation relationships between statements and models the network based on this. In the network, the suspiciousness of statements is used as node weights, and the correlation between statements is used as edge weights. We chose the eigenvector centrality that takes into account the degree centrality of statements and the importance of neighboring statements to calculate the importance of each statement, and used it as a weight to incorporate into the weighted suspiciousness calculation of the statement. Finally, we applied the FL-WARMCN model for experimental validation on the Defects4J dataset. The results showed that the model was significantly superior to other baselines. In addition, we analyzed the impact of different node and edge weights on model performance.

Celastrol alleviates atopic dermatitis by regulating Ezrin-mediated mitochondrial fission and fusion.

Celastrol, a bioactive molecule extracted from the plant Tripterygium wilfordii Hook F., possesses anti-inflammatory, anti-obesity and anti-tumour properties. Despite its efficacy in improving erythema and scaling in psoriatic mice, the specific therapeutic mechanism of celastrol in atopic dermatitis (AD) remains unknown. This study aims to examine the role and mechanism of celastrol in AD using TNF-α-stimulated HaCaT cells and DNCB-induced Balb/c mice as invitro and invivo AD models, respectively. Celastrol was found to inhibit the increased epidermal thickness, reduce spleen and lymph node weights, attenuate inflammatory cell infiltration and mast cell degranulation and decrease thymic stromal lymphopoietin (TSLP) as well as various inflammatory factors (IL-4, IL-13, TNF-α, IL-5, IL-31, IL-33, IgE, TSLP, IL-17, IL-23, IL-1β, CCL11 and CCL17) in AD mice. Additionally, celastrol inhibited Ezrin phosphorylation at Thr567, restored mitochondrial network structure, promoted translocation of Drp1 to the cytoplasm and reduced TNF-α-induced cellular reactive oxygen species (ROS), mitochondrial ROS (mtROS) and mitochondrial membrane potential (MMP) production. Interestingly, Mdivi-1 (a mitochondrial fission inhibitor) and Ezrin-specific siRNAs lowered inflammatory factor levels and restored mitochondrial reticular formation, as well as ROS, mtROS and MMP production. Co-immunoprecipitation revealed that Ezrin interacted with Drp1. Knocking down Ezrin reduced mitochondrial fission protein Drp1 phosphorylation and Fis1 expression while increasing the expression of fusion proteins Mfn1 and Mfn2. The regulation of mitochondrial fission and fusion by Ezrin was confirmed. Overall, celastrol may alleviate AD by regulating Ezrin-mediated mitochondrial fission and fusion, which may become a novel therapeutic reagent for alleviating AD.

A Comparison of Arterial Input Function Interpolation Methods for Patlak Plot Analysis of 68Ga-PSMA-11 PET Prostate Cancer Studies.

Positron emission tomography (PET) imaging enables quantitative assessment of tissue physiology. Dynamic pharmacokinetic analysis of PET images requires accurate estimation of the radiotracer plasma input function to derive meaningful parameter estimates, and small discrepancies in parameter estimation can mimic subtle physiologic tissue variation. This study evaluates the impact of input function interpolation method on the accuracy of Patlak kinetic parameter estimation through simulations modeling the pharmacokinetic properties of [68Ga]-PSMA-11. This study evaluated both trained and untrained methods. Although the mean kinetic parameter accuracy was similar across all interpolation models, the trained node weighting interpolation model estimated accurate kinetic parameters with reduced overall variability relative to standard linear interpolation. Trained node weighting interpolation reduced kinetic parameter estimation variance by a magnitude approximating the underlying physiologic differences between normal and diseased prostatic tissue. Overall, this analysis suggests that trained node weighting improves the reliability of Patlak kinetic parameter estimation for [68Ga]-PSMA-11 PET.

Distance-based dynamic caching and replacement strategy in NDN-IoT networks

The amount of data on the Internet has risen noticeably due to the exponential increase in the number of Internet of Things (IoT) devices. Recently, Information-Centric Networking (ICN) is acknowledged as a suitable framework for creating IoT network protocols. One notable feature of ICN is in-network caching, which stores the data packets on intermediate routers/nodes that can be accessed by the subscribers when required. This work proposes a novel Distance-based dynamic caching and replacement strategy (DBDCRS) which involves a light weight calculation that utilizes the request aggregation to find the request weight of the requests stored in Pending interest table (PIT) and node weight of a node based on the distance from the subscriber and IoT publisher to accomplish the caching decision. To eliminate the data redundancy, the scheme caches a particular content on a single node along the specific interest-data path. Additionally, a custom cache replacement policy is introduced which retains the essential contents in cache. Simulation results show that our proposed DBDCRS scheme outperform other in-network cache placement schemes in terms of cache hit ratio, server hit reduction ratio, average response latency, and energy consumption.

Location and Price Competition on a Uniform Path with Different Pricing Policies

AbstractThis paper models duopolistic competition between an online retailer and a physical store retailer with the online retailer modelled as a firm with uniform delivered pricing policy and the physical store as a firm with a mill pricing policy. Both firms seek to maximize their respective profits through an appropriate choice of location and price. The market is assumed to be given by a “uniform path” - a tree whose node weights and arc lengths are equal. Modelling reality, we consider two alternate types of transportation costs faced by the online retailer: either dependent on the relative location of the firm and a customer or independent of it. Beginning with the framework of a Stackelberg, i.e., sequential game, optimal location and price strategies are analytically derived for both sequences of market entry by the two competitors. Cases in which the leader faces the first entry paradox or can become a monopolist by strategically deterring entry by the follower are delineated. Thereafter, Nash Equilibrium solutions to the simultaneous game are identified. Salient insights from the results include: (a) the competitive pressure faced by the physical store in the presence of online competition (b) the inability of an online retailer to compete in “large” markets under the first type of transportation cost and (c) the advantage to the physical retailer of being a market leader.

Improving Temporal Treemaps by Minimizing Crossings

AbstractTemporal trees are trees that evolve over a discrete set of time steps. Each time step is associated with a node‐weighted rooted tree and consecutive trees change by adding new nodes, removing nodes, splitting nodes, merging nodes, and changing node weights. Recently, two‐dimensional visualizations of temporal trees called temporal treemaps have been proposed, representing the temporal dimension on the x‐axis, and visualizing the tree modifications over time as temporal edges of varying thickness. The tree hierarchy at each time step is depicted as a vertical, one‐dimensional nesting relationships, similarly to standard, non‐temporal treemaps. Naturally, temporal edges can cross in the visualization, decreasing readability. Heuristics were proposed to minimize such crossings in the literature, but a formal characterization and minimization of crossings in temporal treemaps was left open. In this paper, we propose two variants of defining crossings in temporal treemaps that can be combinatorially characterized. For each variant, we propose an exact optimization algorithm based on integer linear programming and heuristics based on graph drawing techniques. In an extensive experimental evaluation, we show that on the one hand the exact algorithms reduce the number of crossings by a factor of 20 on average compared to the previous algorithms. On the other hand, our new heuristics are faster by a factor of more than 100 and still reduce the number of crossings by a factor of almost three.

Neighbors selective Graph Convolutional Network for homophily and heterophily

Graph Convolutional Networks (GCNs) gain remarkable success in graph-related tasks under homophily graph assumption—most connected nodes have the same label. However, this assumption is fragile since heterophily is common in real-world networks, where most linked nodes have different labels. Some methods address heterophily using graph-level weighted combinations of multi-hop neighbor representations. However, they introduce noise and irrelevant information from two aspects: (1) higher-order neighborhoods may include more neighbors with different labels than those with the same label; (2) graph-level weighted combination fails to capture distinct node properties. To address these issues, firstly, a Neighborhood Distribution-induced Similarity (NDS) measure is developed to identify potential similar neighbors for nodes. Secondly, a node-level fusion mechanism Selective-Neighbors Gated Unit (SNGU) is designed to adaptively aggregate potential neighbors, first-hop neighbors, and the node itself by learning distinct weights for nodes. By combining the above two designs, this work proposes a novel Neighbors Selective Graph Convolutional Network (NSGCN), which allows nodes to selectively receive relevant neighbor information for better node representations, effectively modeling homophily and heterophily. Experiments on 10 widely used real-world datasets with varying properties demonstrate the superiority of the NSGCN, surpassing a strong baseline with an absolute increase of 9.83% in classification accuracy. The code is available at https://github.com/GGA23/NSGCN.