Neighborhood Information Research Articles

Hierarchical Graph Capsule Networks for Molecular Function Classification With Disentangled Representations.

In biochemistry, graph structures have been widely used for modeling compounds, proteins, functional interactions, etc. A common task that divides these graphs into different categories, known as graph classification, highly relies on the quality of the representations of graphs. With the advance in graph neural networks, message-passing-based methods are adopted to iteratively aggregate neighborhood information for better graph representations. These methods, though powerful, still suffer from some shortcomings. The first challenge is that pooling-based methods in graph neural networks may sometimes ignore the part-whole hierarchies naturally existing in graph structures. These part-whole relationships are usually valuable for many molecular function prediction tasks. The second challenge is that most existing methods do not take the heterogeneity embedded in graph representations into consideration. Disentangling the heterogeneity will increase the performance and interpretability of models. This paper proposes a graph capsule network for graph classification tasks with disentangled feature representations learned automatically by well-designed algorithms. This method is capable of, on the one hand, decomposing heterogeneous representations to more fine-grained elements, whilst on the other hand, capturing part-whole relationships using capsules. Extensive experiments performed on several public-available biochemistry datasets demonstrated the effectiveness of the proposed method, compared with nine state-of-the-art graph learning methods.

Optimal dispatch for cross-regional integrated energy system with renewable energy uncertainties: A unified spatial-temporal cooperative framework

Cross-regional energy sharing is a promising way to alleviate the imbalance in energy distribution. However, there is a risk that the uncertain fluctuations of intermittent renewable energy sources (RES) could be transmitted across regional transmissions, resulting in large-scale system economic inefficiency and security concerns. To this end, a unified spatial-temporal cooperative framework for the integrated energy system, which considers the coordination between intra-regional multi-energy coupling and cross-regional multi-agent-system energy sharing, is proposed in this paper. First, in the temporal dimension, a multi-time-scale stochastic decision-making model is established based on stochastic model predictive control (SMPC). Thus, slow-response units are arranged to gain economic goals in day-ahead stage and the closed-loop rolling adjustment is implemented on fast-response units to achieve accuracy supply-demand balance in real-time stage. Second, in the spatial dimension, a fully distributed dispatch approach is adopted to address the information sharing barriers, which only requires neighboring information and considers consistency of boundary control variables on border lines. Finally, to improve the computational efficiency, the established complex time sequential model is decomposed into a set of time-discrete parallel sub-problems, using the optimal condition decomposition (OCD) technique. Simulation results on a cross-regional test system demonstrate that proposed framework can increase the wind power consumption capacity and reduce operation cost through cross-regional cooperation, while the parallel decomposition can save 67.25 % of the computation time relative to the serial computation.

DHM-Net: Deep Hypergraph Modeling for Robust Feature Matching.

We present a novel deep hypergraph modeling architecture (called DHM-Net) for feature matching in this paper. Our network focuses on learning reliable correspondences between two sets of initial feature points by establishing a dynamic hypergraph structure that models group-wise relationships and assigns weights to each node. Compared to existing feature matching methods that only consider pair-wise relationships via a simple graph, our dynamic hypergraph is capable of modeling nonlinear higher-order group-wise relationships among correspondences in an interaction capturing and attention representation learning fashion. Specifically, we propose a novel Deep Hypergraph Modeling block, which initializes an overall hypergraph by utilizing neighbor information, and then adopts node-to-hyperedge and hyperedge-to-node strategies to propagate interaction information among correspondences while assigning weights based on hypergraph attention. In addition, we propose a Differentiation Correspondence-Aware Attention mechanism to optimize the hypergraph for promoting representation learning. The proposed mechanism is able to effectively locate the exact position of the object of importance via the correspondence aware encoding and simple feature gating mechanism to distinguish candidates of inliers. In short, we learn such a dynamic hypergraph format that embeds deep group-wise interactions to explicitly infer categories of correspondences. To demonstrate the effectiveness of DHM-Net, we perform extensive experiments on both real-world outdoor and indoor datasets. Particularly, experimental results show that DHM-Net surpasses the state-of-the-art method by a sizable margin. Our approach obtains an 11.65% improvement under error threshold of 5° for relative pose estimation task on YFCC100M dataset. Code will be released at https://github.com/CSX777/DHM-Net.

EBMGC-GNF: Efficient Balanced Multi-view Graph Clustering via Good Neighbor Fusion.

Exploiting consistent structure from multiple graphs is vital for multi-view graph clustering. To achieve this goal, we propose an Efficient Balanced Multi-view Graph Clustering via Good Neighbor Fusion (EBMGC-GNF) model which comprehensively extracts credible consistent neighbor information from multiple views by designing a Cross-view Good Neighbors Voting module. Moreover, a novel balanced regularization term based on p-power function is introduced to adjust the balance property of clusters, which helps the model adapt to data with different distributions. To solve the optimization problem of EBMGC-GNF, we transform EBMGC-GNF into an efficient form with graph coarsening method and optimize it based on accelareted coordinate descent algorithm. In experiments, extensive results demonstrate that, in the majority of scenarios, our proposals outperform state-of-the-art methods in terms of both effectiveness and efficiency.

Neighbor-Guided Pseudo-Label Generation and Refinement for Single-Frame Supervised Temporal Action Localization.

Due to the sparse single-frame annotations, current Single-Frame Temporal Action Localization (SF-TAL) methods generally employ threshold-based pseudo-label generation strategies. However, these approaches suffer from inefficient data utilization, as only parts of unlabeled frames with confidence scores surpassing a predefined threshold are selected for training. Moreover, the variability of single-frame annotations and unreliable model predictions introduce pseudo-label noise. To address these challenges, we propose two strategies by using the relationship of the video segments with their neighbors': 1) temporal neighbor-guided soft pseudo-label generation (TNPG); and 2) semantic neighbor-guided pseudo-label refinement (SNPR). TNPG utilizes a local-global self-attention mechanism in a transformer encoder to capture temporal neighbor information while focusing on the whole video. Then the generated self-attention map is multiplied by the network predictions to propagate information between labeled and unlabeled frames, and produce soft pseudo-label for all segments. Despite this, label noise persists due to unreliable model predictions. To mitigate this, SNPR refines pseudo-labels based on the assumption that predictions should resemble their semantic nearest neighbors'. Specifically, we search for semantic nearest neighbors of each video segment by cosine similarity in the feature space. Then the refined soft pseudo-labels can be obtained by a weight combination of the original pseudo-label and the semantic nearest neighbors'. Finally, the model can be trained with the refined pseudo-labels, and the performance has been greatly improved. Comprehensive experimental results on different benchmarks show that we achieve state-of-the-art performances on THUMOS14, ActivityNet1.2, and ActivityNet1.3 datasets.

Multi-Stage Network with Geometric Semantic Attention for Two-View Correspondence Learning.

The removal of outliers is crucial for establishing correspondence between two images. However, when the proportion of outliers reaches nearly 90%, the task becomes highly challenging. Existing methods face limitations in effectively utilizing geometric transformation consistency (GTC) information and incorporating geometric semantic neighboring information. To address these challenges, we propose a Multi-Stage Geometric Semantic Attention (MSGSA) network. The MSGSA network consists of three key modules: the multi-branch (MB) module, the GTC module, and the geometric semantic attention (GSA) module. The MB module, structured with a multi-branch design, facilitates diverse and robust spatial transformations. The GTC module captures transformation consistency information from the preceding stage. The GSA module categorizes input based on the prior stage's output, enabling efficient extraction of geometric semantic information through a graph-based representation and inter-category information interaction using Transformer. Extensive experiments on the YFCC100M and SUN3D datasets demonstrate that MSGSA outperforms current state-of-the-art methods in outlier removal and camera pose estimation, particularly in scenarios with a high prevalence of outliers. Source code is available at https://github.com/shuyuanlin.

Fuzzy C-means clustering algorithm based on adaptive neighbors information

Fuzzy C-means clustering algorithm based on adaptive neighbors information

Fine-Grained Essential Tensor Learning for Robust Multi-View Spectral Clustering.

Multi-view subspace clustering (MVSC) has drawn significant attention in recent study. In this paper, we propose a novel approach to MVSC. First, the new method is capable of preserving high-order neighbor information of the data, which provides essential and complicated underlying relationships of the data that is not straightforwardly preserved by the first-order neighbors. Second, we design log-based nonconvex approximations to both tensor rank and tensor sparsity, which are effective and more accurate than the convex approximations. For the associated shrinkage problems, we provide elegant theoretical results for the closed-form solutions, for which the convergence is guaranteed by theoretical analysis. Moreover, the new approximations have some interesting properties of shrinkage effects, which are guaranteed by elegant theoretical results. Extensive experimental results confirm the effectiveness of the proposed method.

Current computational methods for spatial transcriptomics in cancer biology.

Cells in multicellular organisms constitute a self-organizing society by interacting with their neighbors. Cancer originates from malfunction of cellular behavior in the context of such a self-organizing system. The identities or characteristics of individual tumor cells can be represented by the hallmark of gene expression or transcriptome, which can be addressed using single-cell dissociation followed by RNA sequencing. However, the dissociation process of single cells results in losing the cellular address in tissue or neighbor information of each tumor cell, which is critical to understanding the malfunctioning cellular behavior in the microenvironment. Spatial transcriptomics technology enables measuring the transcriptome which is tagged by the address within a tissue. However, to understand cellular behavior in a self-organizing society, we need to apply mathematical or statistical methods. Here, we provide a review on current computational methods for spatial transcriptomics in cancer biology.

Parent Grain Reconstruction in an Additive Manufactured Titanium Alloy

Electron backscatter diffraction (EBSD) is an excellent tool for characterizing the crystallographic orientation aspects of the microstructure of polycrystalline material. In some additively manufactured materials, the material may undergo a phase transformation during the forming process. Although EBSD can only characterize the final microstructure, neighbor information from orientation mapping allows the microstructure before the phase transformation to be reconstructed, provided that the parent–child orientation relationship is known. An investigation of the effectiveness of the reconstruction algorithms for capturing the grain size as well as orientation gradients is undertaken with a focus on additively manufactured Ti-alloy. The EBSD results, coupled with reconstruction algorithms, reveal information on the prior grain size as well as the plastic flow of the material.

Unsupervised spatial self-similarity difference-based change detection method for multi-source heterogeneous images

Multi-source heterogeneous change detection has been widely used in dynamic disaster monitoring, land cover updating, etc. Various methods have been proposed to make heterogeneous data comparable. However, heterogeneous images are difficult to compare directly and may be affected by noise. Most existing methods obtain change information through mapping and regression, lacking the utilisation of image spatial information and a comprehensive portrayal of the changes, which may affect change detection results. To address these challenges, we propose an unsupervised spatial self-similarity difference-based change detection (USSD) method for multi-source heterogeneous images to evaluate the similarity of spatial relationships in heterogeneous images. First, the images are divided into image blocks to construct spatial self-difference images between individual image blocks aiming to make the data comparable. Second, the change information is portrayed in terms of both the magnitude differences and similarity differences to obtain a more comprehensive spatial self-difference change magnitude map. Then, the spatial neighbourhood information of the spatial self-difference change magnitude map is considered to avoid noise. Experimental results on six open datasets indicate that the overall accuracy of the USSD method was approximately 85%–95%. This method improves the change magnitude map discrimination, better detects the change region, and avoids noise in synthetic aperture radar images.

MNCLCDA: predicting circRNA-drug sensitivity associations by using mixed neighbourhood information and contrastive learning

BackgroundcircRNAs play an important role in drug resistance and cancer development. Recently, many studies have shown that the expressions of circRNAs in human cells can affect the sensitivity of cells to therapeutic drugs, thus significantly influencing the therapeutic effects of these drugs. Traditional biomedical experiments required to verify this sensitivity relationship are not only time-consuming but also expensive. Hence, the development of an efficient computational approach that can accurately predict the novel associations between drug sensitivities and circRNAs is a crucial and pressing need.MethodsIn this research, we present a novel computational framework called MNCLCDA, which aims to predict the potential associations between drug sensitivities and circRNAs to assist with medical research. First, MNCLCDA quantifies the similarity between the given drug and circRNA using drug structure information, circRNA gene sequence information, and GIP kernel information. Due to the existence of noise in similarity information, we employ a preprocessing approach based on random walk with restart for similarity networks to efficiently capture the useful features of circRNAs and drugs. Second, we use a mixed neighbourhood graph convolutional network to obtain the neighbourhood information of nodes. Then, a graph-based contrastive learning method is used to enhance the robustness of the model, and finally, a double Laplace-regularized least-squares method is used to predict potential circRNA-drug associations through the kernel matrices in the circRNA and drug spaces.ResultsNumerous experimental results show that MNCLCDA outperforms six other advanced methods. In addition, the excellent performance of our proposed model in case studies illustrates that MNCLCDA also has the ability to predict the associations between drug sensitivity and circRNA in practical situations.ConclusionsAfter a large number of experiments, it is illustrated that MNCLCDA is an efficient tool for predicting the potential associations between drug sensitivities and circRNAs, thereby can provide some guidance for clinical trials.

A Distributed Multicast QoS Routing Construction Approach in Information-Centric Networking

Many applications suitable for multicast transmission, such as video conferencing and live e-commerce, demand high Quality of Service (QoS) and require data delivery to be completed within specified delay constraints. Some methods have been proposed for constructing delay-constrained multicast routing based on network state. However, obtaining precise network latency can be challenging, resulting in inaccuracies in delay-constrained routing calculations and, ultimately, the inability to meet application requirements. Additionally, many methods engage in an indiscriminate exploration of potential paths in the network, causing significant message processing overhead. This paper proposes an Information-Centric Networking (ICN)-based approach for delay-constrained multicast routing. Our method dynamically constructs multicast paths from tree nodes to receivers based on real-time path status detection during the join message propagation phase. Additionally, we present a method for acquiring neighborhood state information to facilitate real-time routing decisions. To curtail indiscriminate path exploration, our approach uses the ICN Name Resolution System (NRS) to obtain and select potential optimal tree nodes. For this purpose, we design a multicast service registration and resolution mechanism using the ICN Name Resolution System (NRS). Simulation results indicate that our approach exhibits a higher success ratio and concurrently incurs lower message processing overhead than some other methods, particularly in situations with stringent delay constraints.

U-Transformer-based multi-levels refinement for weakly supervised action segmentation

Action segmentation is a research hotspot in human action analysis, which aims to split videos into segments of different actions. Recent algorithms have achieved great success in modeling based on temporal convolution, but these methods weight local or global timing information through additional modules, ignoring the existing long-term and short-term information connections between actions. This paper proposes a U-Transformer structure based on multi-level refinement, introduces neighborhood attention to learn the neighborhood information of adjacent frames, and aggregates video frame features to effectively process long-term sequence information. Then a loss optimization strategy is proposed to smooth the original classification effect and generate a more accurate calibration sequence by introducing a pairing similarity optimization method based on deep feature learning. In addition, we propose a timestamp supervised training method to generate complete information for actions based on pseudo-label predictions for action boundary predictions. Experiments on three challenging action segmentation datasets, 50Salads, GTEA, and Breakfast, show that our model performs state-of-the-art models, and our weakly supervised model also performs comparably to fully supervised performance.

Detection of E-Commerce Fraud Review via Self-Paced Graph Contrast Learning

Abstract Recently, graph neural networks (GNNs) have been widely used for e-commerce review fraud detection by aggregating the neighborhood information of nodes in various relationships to highlight the suspiciousness of nodes. However, existing GNN-based detection methods are susceptible to sample class imbalance and fraud camouflage problems, resulting in poor quality of constructed graph structures and inability to learn reliable node embeddings. To address the above problems, we propose a novel e-commerce review fraud detection method based on self-paced graph contrast learning (SPCL-GNN). Firstly, the method constructs a subgraph by initially selecting nodes through a labeled balanced extractor. Secondly, the subgraph connections are filtered and complemented by combining self-paced graph contrast learning and an adaptive neighbor sampler to obtain an optimized graph structure. Again, an attention mechanism is introduced in intra- and inter-relationship aggregation to focus on the importance of aggregation under different relationships. Finally, the quality of the node embedding representation is further improved by maximizing the mutual information between the local and global representations. Experimental results on the Amazon and YelpChi datasets show that SPCL-GNN significantly outperforms the baseline.

Shared neighbors rough set model and neighborhood classifiers

Neighborhood relations are established by employing a distance metric and parameters, where the distance metric is utilized to measure the dissimilarities between objects. The information about adjacent objects tends to vary, even though these objects belong to a neighborhood. Taking into account neighborhood information, this paper introduces a similarity measure of shared neighbors, which is determined by the overlapping neighbors in a neighborhood relation. The similarity measure, induced by a neighborhood relation, is superior to the distance metric in quantifying differences, and it is crucial to note that the neighborhood relation can be any binary relation. A neighborhood of shared neighbors is proposed based on the proposed similarity measure, which consists of the most similar neighbors. Subsequently, we propose the shared neighbors rough set model, which is regarded as a framework of fusing neighborhood information. Moreover, an attribute reduction algorithm and a neighborhood classifier of shared neighbors are designed. The experimental results demonstrate the effectiveness of the neighborhood classifier and the shared neighbors rough set model.

A Fault Detection and Isolation Method via Shared Nearest Neighbor for Circulating Fluidized Bed Boiler

Accurate and timely fault detection and isolation (FDI) improve the availability, safety, and reliability of target systems and enable cost-effective operations. In this study, a shared nearest neighbor (SNN)-based method is proposed to identify the fault variables of a circulating fluidized bed boiler. SNN is a derivative method of the k-nearest neighbor (kNN), which utilizes shared neighbor information. The distance information between these neighbors can be applied to FDI. In particular, the proposed method can effectively detect faults by weighing the distance values based on the number of neighbors they share, thereby readjusting the distance values based on the shared neighbors. Moreover, the data distribution is not constrained; therefore, it can be applied to various processes. Unlike principal component analysis and independent component analysis, which are widely used to identify fault variables, the main advantage of SNN is that it does not suffer from smearing effects, because it calculates the contributions from the original input space. The proposed method is applied to two case studies and to the failure case of a real circulating fluidized bed boiler to confirm its effectiveness. The results show that the proposed method can detect faults earlier (1 h 39 min 46 s) and identify fault variables more effectively than conventional methods.

A FULLY CONNECTED CHANGE DETECTION METHOD OF SAR IMAGES FUSING ORIGINAL IMAGE FEATURES AND CHANGE DETECTION RESULTS

Abstract. The primary strategy to eliminate the effect of scatter noise in synthetic aperture radar (SAR) imagery is usually through filtering or combining neighborhood information. However, both approaches to reducing noise reduce the detection accuracy of change edges with similar characteristics to scatter noise points. Considering the above problems, this letter proposes a post-processing method that applies a fully connected conditional random field theoretical model to fuse the original image information with the initial change detection results. The method first takes the original image information and the initial change detection results as a priori conditions. Secondly, the global spatial information in the original image and the label values in the initial change detection results are fully considered when detecting the changed and unchanged pixels to establish a fully connected relationship between all the pixels and find the label distribution probability of each pixel under the condition of noise suppression, and finally obtain better change detection results. The experimental results on the real SAR dataset confirm the proposed method's effectiveness, robustness, and efficiency.

Spectral clustering and embedding with inter-class topology-preserving

Spectral clustering and embedding are critical in data mining and rely heavily on pre-computed graph structure quality. Some methods simultaneously learn graph structures and low-dimensional embeddings to achieve high-quality graph structures with separable categories. However, existing approaches prioritize inter-class separability while overlooking the inter-class topological structure. Consequently, the learned low-dimensional manifold representations may fail to capture inter-class semantic information, causing the learned graph structures to represent categories partitioning inaccurately. To address these issues, we come up with the inter-class topology and utilize category anchors to indicate class locations in low-dimensional space. The inter-class topology reflects the elaborate structural neighbor information of class, which makes the neighbor classes rationally separable and helps to learn better low-dimensional embedding representations. The spectral embedded clustering method and some variants are proposed to preserve inter-class topology while making categories separable. Furthermore, we ensure intra-class sample compactness in low-dimensional space by assigning sample embeddings near the anchors. We analyze the importance of inter-class topology preserving and intra-class compactness learning. Extensive experiments on publicly available datasets demonstrate that our method is competitive with popular graph clustering and embedding baselines and effectively captures the inter-class topology.

EEI-NET: EDGE-ENHANCED INTERPOLATION NETWORK FOR SEMANTIC SEGMENTATION OF HISTORICAL BUILDING POINT CLOUDS

Abstract. In recent years, the conservation research of historical buildings and cultural relics has received a lot of attention from the state and the people, which not only provides a deeper understanding of their historical value and cultural significance, but also promotes the expansion of conservation research to the three-dimensional level. In this context, the semantic segmentation of historical building components is particularly important, which can provide basic support for various historical building applications, such as research and study of historical buildings, repair and protection, and 3D fine reconstruction, etc. However, most of the current methods for semantic segmentation of point clouds of historical buildings suffer from the problems of not being able to fully exploit the local neighborhood information of point clouds and poor edge segmentation. Therefore, we propose a new deep learning semantic segmentation-based approach, which we call EEI-Net. It is an end-to-end deep neural network in which we designed an edge enhancement interpolation (EEI) module and an edge interaction classifier (EIC). The edge enhancement interpolation module performs edge enhancement interpolation by fusing multi-layer features between the encoder and decoder. The edge interaction classifier enables the interaction of edge information through information transfer between individual nodes. EEI-Net incorporates contextual features and better preserves and enhances the edge information of the point cloud. We conduct experiments on the constructed historical architecture dataset, and the results show that the proposed EEI-Net has better performance.