Raw Features Research Articles

MOGAT: A Multi-Omics Integration Framework Using Graph Attention Networks for Cancer Subtype Prediction.

Accurate cancer subtype prediction is crucial for personalized medicine. Integrating multi-omics data represents a viable approach to comprehending the intricate pathophysiology of complex diseases like cancer. Conventional machine learning techniques are not ideal for analyzing the complex interrelationships among different categories of omics data. Numerous models have been suggested using graph-based learning to uncover veiled representations and network formations unique to distinct types of omics data to heighten predictions regarding cancers and characterize patients' profiles, amongst other applications aimed at improving disease management in medical research. The existing graph-based state-of-the-art multi-omics integration approaches for cancer subtype prediction, MOGONET, and SUPREME, use a graph convolutional network (GCN), which fails to consider the level of importance of neighboring nodes on a particular node. To address this gap, we hypothesize that paying attention to each neighbor or providing appropriate weights to neighbors based on their importance might improve the cancer subtype prediction. The natural choice to determine the importance of each neighbor of a node in a graph is to explore the graph attention network (GAT). Here, we propose MOGAT, a novel multi-omics integration approach, leveraging GAT models that incorporate graph-based learning with an attention mechanism. MOGAT utilizes a multi-head attention mechanism to extract appropriate information for a specific sample by assigning unique attention coefficients to neighboring samples. Based on our knowledge, our group is the first to explore GAT in multi-omics integration for cancer subtype prediction. To evaluate the performance of MOGAT in predicting cancer subtypes, we explored two sets of breast cancer data from TCGA and METABRIC. Our proposed approach, MOGAT, outperforms MOGONET by 32% to 46% and SUPREME by 2% to 16% in cancer subtype prediction in different scenarios, supporting our hypothesis. Our results also showed that GAT embeddings provide a better prognosis in differentiating the high-risk group from the low-risk group than raw features.

AFRNet: Anchor-Free Object Detection Using Roadside LiDAR in Urban Scenes

In urban settings, roadside infrastructure LiDAR is a ground-based remote sensing system that collects 3D sparse point clouds for the traffic object detection of vehicles, pedestrians, and cyclists. Current anchor-free algorithms for 3D point cloud object detection based on roadside infrastructure face challenges related to inadequate feature extraction, disregard for spatial information in large 3D scenes, and inaccurate object detection. In this study, we propose AFRNet, a two-stage anchor-free detection network, to address the aforementioned challenges. We propose a 3D feature extraction backbone based on the large sparse kernel convolution (LSKC) feature set abstraction module, and incorporate the CBAM attention mechanism to enhance the large scene feature extraction capability and the representation of the point cloud features, enabling the network to prioritize the object of interest. After completing the first stage of center-based prediction, we propose a refinement method based on attentional feature fusion, where fused features incorporating raw point cloud features, voxel features, BEV features, and key point features are used for the second stage of refinement to complete the detection of 3D objects. To evaluate the performance of our detection algorithms, we conducted experiments using roadside LiDAR data from the urban traffic dataset DAIR-V2X, based on the Beijing High-Level Automated Driving Demonstration Area. The experimental results show that AFRNet has an average of 5.27 percent higher detection accuracy than CenterPoint for traffic objects. Comparative tests further confirm that our method achieves high accuracy in roadside LiDAR object detection.

Mouse Dynamics Behavioral Biometrics: A Survey

Utilization of the Internet in our everyday lives has made us vulnerable in terms of privacy and security of our data and systems. Therefore, there is a pressing need to protect our data and systems by improving authentication mechanisms, which are expected to be low cost, unobtrusive, and ideally ubiquitous in nature. Behavioral biometric modalities such as mouse dynamics (mouse behaviors on a graphical user interface (GUI)) and widget interactions (another modality closely related to mouse dynamics that also considers the target (widget) of a GUI interaction, such as links, buttons, and combo-boxes) can bolster the security of existing authentication systems because of their ability to distinguish individuals based on their unique features. As a result, it can be difficult for an imposter to impersonate these behavioral biometrics, making them suitable for authentication. In this article, we survey the literature on mouse dynamics and widget interactions dated from 1897 to 2023. We begin our survey with an account of the psychological perspectives on behavioral biometrics. We then analyze the literature along the following dimensions: tasks and experimental settings for data collection, taxonomy of raw attributes, feature extractions and mathematical definitions, publicly available datasets, algorithms (statistical, machine learning, and deep learning), data fusion, performance, and limitations. We end the paper with presenting challenges and promising research opportunities.

Implementing Graph-Theoretic Feature Selection by Quantum Approximate Optimization Algorithm.

Feature selection plays a significant role in computer science; nevertheless, this task is intractable since its search space scales exponentially with the number of dimensions. Motivated by the potential advantages of near-term quantum computing, three graph-theoretic feature selection (GTFS) methods, including minimum cut (MinCut)-based, densest k -subgraph (DkS)-based, and maximal-independent set/minimal vertex cover (MIS/MVC)-based, are investigated in this article, where the original graph-theoretic problems are naturally formulated as the quadratic problems in binary variables and then solved using the quantum approximate optimization algorithm (QAOA). Specifically, three separate graphs are created from the raw feature set, where the vertex set consists of individual features and pairwise measure describes the edge. The corresponding feature subset is generated by deriving a subgraph from the established graph using QAOA. For the above three GTFS approaches, the solving procedure and quantum circuit for the corresponding graph-theoretic problems are formulated with the framework of QAOA. In addition, those proposals could be employed as a local solver and integrated with the Tabu search algorithm for solving large-scale GTFS problems utilizing limited quantum bit resource. Finally, extensive numerical experiments are conducted with 20 publicly available datasets and the results demonstrate that each model is superior to its classical scheme. In addition, the complexity of each model is only O(p n2) even in the worst cases, where p is the number of layers in QAOA and n is the number of features.

FreshGNN: Reducing Memory Access via Stable Historical Embeddings for Graph Neural Network Training

A key performance bottleneck when training graph neural network (GNN) models on large, real-world graphs is loading node features onto a GPU. Due to limited GPU memory, expensive data movement is necessary to facilitate the storage of these features on alternative devices with slower access (e.g. CPU memory). Moreover, the irregularity of graph structures contributes to poor data locality which further exacerbates the problem. Consequently, existing frameworks capable of efficiently training large GNN models usually incur a significant accuracy degradation because of the currently-available shortcuts involved. To address these limitations, we instead propose FreshGNN, a general-purpose GNN mini-batch training framework that leverages a historical cache for storing and reusing GNN node embeddings instead of re-computing them through fetching raw features at every iteration. Critical to its success, the corresponding cache policy is designed, using a combination of gradient-based and staleness criteria, to selectively screen those embeddings which are relatively stable and can be cached, from those that need to be re-computed to reduce estimation errors and subsequent downstream accuracy loss. When paired with complementary system enhancements to support this selective historical cache, FreshGNN is able to accelerate the training speed on large graph datasets such as ogbn-papers100M and MAG240M by 3.4× up to 20.5× and reduce the memory access by 59%, with less than 1% influence on test accuracy.

Fault Diagnosis in Centrifugal Pumps: A Dual-Scalogram Approach with Convolution Autoencoder and Artificial Neural Network

This paper proposes a new fault diagnosis method for centrifugal pumps by combining signal processing with deep learning techniques. Centrifugal pumps facilitate fluid transport through the energy generated by the impeller. Throughout the operation, variations in the fluid pressure at the pump’s inlet may impact the generalization of traditional machine learning models trained on raw statistical features. To address this concern, first, vibration signals are collected from centrifugal pumps, followed by the application of a lowpass filter to isolate frequencies indicative of faults. These signals are then subjected to a continuous wavelet transform and Stockwell transform, generating two distinct time–frequency scalograms. The Sobel filter is employed to further highlight essential features within these scalograms. For feature extraction, this approach employs two parallel convolutional autoencoders, each tailored for a specific scalogram type. Subsequently, extracted features are merged into a unified feature pool, which forms the basis for training a two-layer artificial neural network, with the aim of achieving accurate fault classification. The proposed method is validated using three distinct datasets obtained from the centrifugal pump under varying inlet fluid pressures. The results demonstrate classification accuracies of 100%, 99.2%, and 98.8% for each dataset, surpassing the accuracies achieved by the reference comparison methods.

Criteria for Identifying Knapping Skill Level Through the Analysis of Lithic Cores: An Example from Val Lastari, Late Palaeolithic, Italy

ABSTRACT The application of knapping skill analysis has increased over the years, starting from the 1980s and, by now, being employed all over the world on different material cultures and chronological periods. We studied the cores reduced at Val Lastari, a Recent Epigravettian lithic workshop in north-eastern Italy, for recognizing the most influential technological variables to define stoneknapping behaviors. In our case, knapping accidents and cores appearance had more weight than criteria such as core preparation and raw material features thanks to the strategies of exploitation chosen by hunter-gatherers of Val Lastari, and to the ecological context they interacted with. This is the first time that a study of its kind has been proposed for an Italian site, becoming a unique possibility to enrich our knowledge about Epigravettian social learning and organization with new and stimulating information.

Zwitterion mediated anti-protein adsorption on polypropylene mesh to reduce inflammation for efficient hernia repair

The effectiveness of polypropylene (PP) mesh is often compromised by severe inflammation. Engineering anti-inflammatory coatings has significant implications for PP mesh to repair unwanted hernias. Here, we presented a facile strategy to develop an anti-fouling coating consisting of zwitterionic poly(carboxybetaine methacrylate) (PCBMA), which could prohibit protein adsorption to endow PP mesh with anti-inflammatory efficacy. The incorporation of PCBMA coating had little impact on the raw features of PP mesh. While the modified mesh PCBMA-PP possessed noticeable hydrophilicity increase and surface charge reduction. The excellent lubricity and surface stability enabled PCBMA-PP to exhibit superior anti-fouling capacity, thus efficiently inhibiting the adsorption of proteins. In vivo experiments showed that incorporating the PCBMA layer could provide PP meshes with outstanding anti-inflammatory effects and tissue compatibility for repairing hernias.

Towards vibration-based damage detection of civil engineering structures: overview, challenges, and future prospects

AbstractIn this paper, we delve into the evolving landscape of vibration-based structural damage detection (SDD) methodologies, emphasizing the pivotal role civil structures play in society's wellbeing and progress. While the significance of monitoring the resilience, durability, and overall health of these structures remains paramount, the methodology employed is continually evolving. Our focus encompasses not just the transformation brought by the advent of artificial intelligence but also the nuanced challenges and future directions that emerge from this integration. We shed light on the inherent nonlinearities civil engineering structures face, the limitations of current validation metrics, and the conundrums introduced by inverse analysis. Highlighting machine learning's (ML) transformative role, we discuss how techniques such as artificial neural networks and support vector machine's have expanded the SDD's scope. Deep learning's (DL) contributions, especially the innovative capabilities of convolutional neural network in raw data feature extraction, are elaborated upon, juxtaposed with the potential pitfalls, like data overfitting. We propose future avenues for the field, such as blending undamaged real-world data with simulated damage scenarios and a tilt towards unsupervised algorithms. By synthesizing these insights, our review offers an updated perspective on the amalgamation of traditional SDD techniques with ML and DL, underlining their potential in fostering more robust civil infrastructures.

Girls gone wild: animality, female teenagers, and disidentification in contemporary European cinema

ABSTRACT Contemporary European cinema insistently depicts female teenagers in association with non-human animals. This article focuses on a group of films in which the teenage protagonists approach animality while closely surrounded by systems of exploitation of animals. The films Axolotl Overkill (Helene Hegemann, 2017), Raw (Julia Ducournau, 2016), and Eastern (Piotr Adamski, 2019) feature female teenagers that experience oppression over their bodies. Animality appears both as part of their oppression and as a subversive alternative to negotiate their appointed social role as girls. The analyzed films explore narrative and aesthetic modes of disruption of the division between human/animal, as they articulate constructive encounters and even identification with the human and the non-human. Additionally, these films confront normative constructions of femininity through abject animality. Instead of ‘becoming women’, the protagonists embrace a more fluid transformation. These films propose ways of representing animals without participating in the traditional, spectacularizing gaze that reinforces the hierarchy of dominance of the human over the non-human. In this sense, this article contributes to the aim of animal (film) studies to rethink how visual culture represents the non-human. Lastly, these films appear more broadly as part of a larger tendency in contemporary European cinema inscribing animals in a critique of conventional femininity.

Iterative Deep Structural Graph Contrast Clustering for Multiview Raw Data.

Multiview clustering has attracted increasing attention to automatically divide instances into various groups without manual annotations. Traditional shadow methods discover the internal structure of data, while deep multiview clustering (DMVC) utilizes neural networks with clustering-friendly data embeddings. Although both of them achieve impressive performance in practical applications, we find that the former heavily relies on the quality of raw features, while the latter ignores the structure information of data. To address the above issue, we propose a novel method termed iterative deep structural graph contrast clustering (IDSGCC) for multiview raw data consisting of topology learning (TL), representation learning (RL), and graph structure contrastive learning to achieve better performance. The TL module aims to obtain a structured global graph with constraint structural information and then guides the RL to preserve the structural information. In the RL module, graph convolutional network (GCN) takes the global structural graph and raw features as inputs to aggregate the samples of the same cluster and keep the samples of different clusters away. Unlike previous methods performing contrastive learning at the representation level of the samples, in the graph contrastive learning module, we conduct contrastive learning at the graph structure level by imposing a regularization term on the similarity matrix. The credible neighbors of the samples are constructed as positive pairs through the credible graph, and other samples are constructed as negative pairs. The three modules promote each other and finally obtain clustering-friendly embedding. Also, we set up an iterative update mechanism to update the topology to obtain a more credible topology. Impressive clustering results are obtained through the iterative mechanism. Comparative experiments on eight multiview datasets show that our model outperforms the state-of-the-art traditional and deep clustering competitors.

An Ensemble Model Health Care Monitoring System.

Internet of things (IoT) is utilized to enhance conventional health care systems in several ways, including patient's disease monitoring. The data gathered by IoT devices is very beneficial to medical facilities and patients. The data needs to be secured against unauthorized modifications because of security and privacy concerns. Conversely, a variety of procedures are offered by block chain technology to safeguard data against modifications. Block chain-based IoT-based health care monitoring is thus a fascinating technical advancement that may aid in easing security and privacy problems associated withthe collection of data during patient monitoring. In this work, we present an ensemble classification-based monitoring system with a block-chain as the foundation for an IoT health care model. Initially, data generation is done by considering the diseases including chronic obstructive pulmonary disease (COPD), lung cancer, and heart disease. The IoT health care data is then preprocessed using enhanced scalar normalization. The preprocessed data was used to extract features such as mutual information (MI), statistical features, adjusted entropy, and raw features. The total classified result is obtained by averaging deep maxout, improved deep convolutional network (IDCNN), and deep belief network (DBN) ensemble classification. Finally, decision-making is done by doctors to suggest treatment based on the classified results from the ensemble classifier. The ensemble model scored the greatest accuracy (95.56%) with accurate disease classification at a learning percentage of 60% compared to traditional classifiers such as neural network (NN) (89.08%), long short term memory (LSTM) (80.63%), deep belief network (DBN) (79.78%) and GT based BSS algorithm (89.08%).

Robust Remote Photoplethysmography Estimation With Environmental Noise Disentanglement.

Remote Photoplethysmography (rPPG) has been attracting increasing attention due to its potential in a wide range of application scenarios such as physical training, clinical monitoring, and face anti-spoofing. On top of conventional solutions, deep-learning approach starts to dominate in rPPG estimation and achieves top-level performance. However, most of them try to integrate preprocessing steps such as the ROI selection into an end-to-end network, which may diverge the attention and also limit the generalization in other scenarios with different input skin regions. In this work, we focus on learning the intrinsic rPPG feature and design a lightweight but effective rPPG estimation network based on spatiotemporal convolution. To further improve the robustness, on top of the basic design we propose the Noise-Disentangled DeeprPPG (ND-DeeprPPG) by disentangling the environmental noise from the raw rPPG feature with an adversarial canonical correlation analysis learning strategy. Background regions are employed as a reference to guide the noise disentangling in a self-supervised manner. Extensive experiments show that our ND-DeeprPPG not only outperforms the state-of-the-arts on heart rate estimation but also exhibits promising robustness in cross-skin-region, cross-dataset scenarios and other rPPG-based tasks.

Sanitized clustering against confounding bias

AbstractReal-world datasets inevitably contain biases that arise from different sources or conditions during data collection. Consequently, such inconsistency itself acts as a confounding factor that disturbs the cluster analysis. Existing methods eliminate the biases by projecting data onto the orthogonal complement of the subspace expanded by the confounding factor before clustering. Therein, the interested clustering factor and the confounding factor are coarsely considered in the raw feature space, where the correlation between the data and the confounding factor is ideally assumed to be linear for convenient solutions. These approaches are thus limited in scope as the data in real applications is usually complex and non-linearly correlated with the confounding factor. This paper presents a new clustering framework named Sanitized Clustering Against confounding Bias, which removes the confounding factor in the semantic latent space of complex data through a non-linear dependence measure. To be specific, we eliminate the bias information in the latent space by minimizing the mutual information between the confounding factor and the latent representation delivered by variational auto-encoder. Meanwhile, a clustering module is introduced to cluster over the purified latent representations. Extensive experiments on complex datasets demonstrate that our SCAB achieves a significant gain in clustering performance by removing the confounding bias.

Fast and Accurate Hyperspectral Image Classification with Window Shape Adaptive Singular Spectrum Analysis

Hyperspectral classification is a task of significant importance in the field of remote sensing image processing, with attaining high precision and rapid classification increasingly becoming a research focus. The classification accuracy depends on the degree of raw HSI feature extraction, and the use of endless classification methods has led to an increase in computational complexity. To achieve high accuracy and fast classification, this study analyzes the inherent features of HSI and proposes a novel spectral–spatial feature extraction method called window shape adaptive singular spectrum analysis (WSA-SSA) to reduce the computational complexity of feature extraction. This method combines similar pixels in the neighborhood to reconstruct every pixel in the window, and the main steps are as follows: rearranging the spectral vectors in the irregularly shaped region, constructing an extended trajectory matrix, and extracting the local spatial and spectral information while removing the noise. The results indicate that, given the small sample sizes in the Indian Pines dataset, the Pavia University dataset, and the Salinas dataset, the proposed algorithm achieves classification accuracies of 97.56%, 98.34%, and 99.77%, respectively. The classification speed is more than ten times better than that of other methods, and a classification time of only about 1–2 s is needed.

Introducing packet-level analysis in programmable data planes to advance Network Intrusion Detection

Programmable data planes offer precise control over the low-level processing steps applied to network packets, serving as a valuable tool for analysing malicious flows in the field of intrusion detection. Albeit with limitations on physical resources and capabilities, they allow for the efficient extraction of detailed traffic information, which can then be utilised by Machine Learning (ML) algorithms responsible for identifying security threats. In addressing resource constraints, existing solutions in the literature rely on compressing network data through the collection of statistical traffic features in the data plane. While this compression saves memory resources in switches and minimises the burden on the control channel between the data and the control plane, it also results in a loss of information available to the Network Intrusion Detection System (NIDS), limiting access to packet payload, categorical features, and the semantic understanding of network communications, such as the behaviour of packets within traffic flows. This paper proposes P4DDLe, a framework that exploits the flexibility of P4-based programmable data planes for packet-level feature extraction and pre-processing. P4DDLe leverages the programmable data plane to extract raw packet features from the network traffic, categorical features included, and to organise them in a way that the semantics of traffic flows are preserved. To minimise memory and control channel overheads, P4DDLe selectively processes and filters packet-level data, so that only the features required by the NIDS are collected. The experimental evaluation with recent Distributed Denial of Service (DDoS) attack data demonstrates that the proposed approach is very efficient in collecting compact and high-quality representations of network flows, ensuring precise detection of DDoS attacks.

SAE3D: Set Abstraction Enhancement Network for 3D Object Detection Based Distance Features.

With the increasing demand from unmanned driving and robotics, more attention has been paid to point-cloud-based 3D object accurate detection technology. However, due to the sparseness and irregularity of the point cloud, the most critical problem is how to utilize the relevant features more efficiently. In this paper, we proposed a point-based object detection enhancement network to improve the detection accuracy in the 3D scenes understanding based on the distance features. Firstly, the distance features are extracted from the raw point sets and fused with the raw features regarding reflectivity of the point cloud to maximize the use of information in the point cloud. Secondly, we enhanced the distance features and raw features, which we collectively refer to as self-features of the key points, in set abstraction (SA) layers with the self-attention mechanism, so that the foreground points can be better distinguished from the background points. Finally, we revised the group aggregation module in SA layers to enhance the feature aggregation effect of key points. We conducted experiments on the KITTI dataset and nuScenes dataset and the results show that the enhancement method proposed in this paper has excellent performance.

Significance of handcrafted features in human activity recognition with attention-based RNN models

Sensors incorporated in devices are a source of temporal data that can be interpreted to learn the context of a user. The smartphone accelerometer sensor generates data streams that form distinct patterns in response to user activities. The human context can be predicted using deep learning models built from raw sensor data or features retrieved from raw data. This study analyzes data streams from the UCI-HAR public dataset for activity recognition to determine 31 handcrafted features in the temporal and frequency domain. Various stacked and combination RNN models, trained with attention mechanisms, are designed to work with computed features. Attention gave the models a good fit. When trained with all features, the two-stacked GRU model performed best with 99% accuracy. Selecting the most promising features helps reduce training time without compromising accuracy. The ranking supplied by the permutation feature importance measure and Shapley values are utilized to identify the best features from the highly correlated features. Models trained using optimal features, as determined by the importance measures, had a 96% accuracy rate. Misclassification in attention-based classifiers occurs in the prediction of dynamic activities, such as walking upstairs and walking downstairs, and in sedentary activities, such as sitting and standing, due to the similar range of each activity’s axis values. Our research emphasizes the design of streamlined neural network architectures, characterized by fewer layers and a reduced number of neurons when compared to existing models in the field, to design lightweight models to be implemented in resource-constraint gadgets.

Distraction-level recognition based on stacking ensemble learning for IVIS secondary tasks

Driver distraction-level recognition while performing secondary tasks in full-touch in-vehicle information systems (FTIVISs) is essential for the harmonious co-driving of human and intelligent vehicle systems. However, there has been little research on this topic. To respond to this issue, this paper proposes a distraction-level recognition framework with a combination of semi-supervised learning, unsupervised learning, and supervised learning. First, unsupervised learning and semi-supervised is used to divide the collected unlabeled samples of driving distraction behavior into three categories of distraction levels: high, medium, and low. Second, the factors influencing the distraction level are explored through a mixed model analysis. Finally, a stacking-based ensemble learning model is proposed to recognize the driver distraction level by supervised learning, with the influencing factors of the distraction level used as model input parameters. To improve the recognition performance of the stacking model, four heterogeneous classifiers selected as the base classifiers, and principal component analysis (PCA) is introduced in the base classifier layer, which improves the ability of the meta-classifier to process raw features based on overfitting resistance. We conducted a real road experiment under different road and FTIVIS task conditions, and the proposed model performed better than traditional machine learning models. In addition, the model exhibited the greatest advantage when the deep neural network (DNN) algorithm was used as the meta-classifier of the model, with a recognition accuracy of 92.5%. The study findings are significant for developing a human–machine co-driving control strategy and improving vehicle driving safety.

Adaptive Graph Convolution Methods for Attributed Graph Clustering

Attributed graph clustering is a challenging task as it requires to jointly model graph structure and node attributes. Although recent advances in graph convolutional networks have shown the effectiveness of graph convolution in combining structural and content information, there is limited understanding of how to properly apply it for attributed graph clustering. Previous methods commonly use a fixed and low order graph convolution, which only aggregates information of few-hop neighbours and hence cannot fully capture the cluster structures of diverse graphs. In this paper, we first propose an adaptive graph convolution method (AGC) for attributed graph clustering, which exploits high-order graph convolutions to capture global cluster structures and adaptively selects an appropriate order <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$k$</tex-math></inline-formula> via intra-cluster distance. While AGC can find a reasonable <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$k$</tex-math></inline-formula> and avoid over-smoothing, it is not sensitive to the gradual decline of clustering performance as <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$k$</tex-math></inline-formula> increases. To search for a better <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$k$</tex-math></inline-formula> , we further propose an improved adaptive graph convolution method (IAGC) that not only observes the variation of intra-cluster distance, but also considers the inconsistencies of filtered features with graph structure and raw features, respectively. We establish the validity of our methods by theoretical analysis and extensive experiments on various benchmark datasets.