Hierarchical Graph Research Articles

Hierarchical Graph Neural Networks with Scale-Aware Readout for Image Classification

This work addresses the importance of incorporating multi-scale information in image representation by proposing a novel approach utilizing hierarchical segmentation and graph neural networks (GNNs). The proposed model, named Hierarchical Image Graph with Scale Importance (HIGSI), leverages hierarchical segmentation to construct graphs that capture relationships between nodes across different scales. This multi-scale representation simultaneously captures intricate details and global context, leading to a richer understanding of image structure than traditional methods. Additionally, a novel Region Graph Readout (RGR) function is introduced to assess the significance of each scale within the graph representation. By combining this multi-scale representation and the RGR function, HIGSI achieves competitive performance on image classification tasks, using smaller graphs or having fewer parameters than existing methods. This work also presents a comparative study with another hierarchical approach and an assessment of HIGSI’s components to investigate its decision-making process and its components’ contribution to the overall performance.

Missile Fault Detection and Localization Based on HBOS and Hierarchical Signed Directed Graph

The rudder surfaces and lifting surfaces of a missile are utilized to acquire aerodynamic forces and moments, adjust the missile’s attitude, and achieve precise strike missions. However, the harsh flying conditions of missiles make the rudder surfaces and lifting surfaces susceptible to faults. In practical scenarios, there is often a scarcity of fault data, and sometimes, it is even difficult to obtain such data. Currently, data-driven fault detection and localization methods heavily rely on fault data, posing challenges for their applicability. To address this issue, this paper proposes an HBOS (Histogram-Based Outlier Score) online fault-detection method based on statistical distribution. This method generates a fault-detection model by fitting the probability distribution of normal data and incorporates an adaptive threshold to achieve real-time fault detection. Furthermore, this paper abstracts the interrelationships between the missile’s flight states and the propagation mechanism of faults into a hierarchical directed graph model. By utilizing bilateral adaptive thresholds, it captures the first fault features of each sub-node and determines the fault propagation effectiveness of each layer node based on the compatibility path principle, thus establishing a fault inference and localization model. The results of semi-physical simulation experiments demonstrate that the proposed algorithm is independent of fault data and exhibits high real-time performance. In multiple sets of simulated tests with randomly parameterized deviations, the fault-detection accuracy exceeds 98% with a false-alarm rate of no more than 0.31%. The fault-localization algorithm achieves an accuracy rate of no less than 97.91%.

Auto-weighted multi-view clustering via hierarchical bipartite graph

The expensive time consumption and hyper-parameters are two main drawbacks of most existing multi-view graph clustering methods. Especially in large-scale data clustering, these two defects are more serious. Aiming at these two problems, we propose a novel auto-weighted multi-view clustering method based on the hierarchical bipartite graph to effectively address these two limitations. Similar to the idea of an anchor graph, firstly, the bisecting k-means method is used instead of traditional method to generate a hierarchical anchor points set. And then, the hierarchical bipartite graph can be constructed between the original points and the anchor points of last layer. Since we only consider the anchor points on the last layer rather than using a larger number of anchors to obtain the bipartite graph, our proposed method will greatly reduce the time consumption. Moreover, the automatic learning strategy was adopted to select the appropriate weights for each view in our proposed approach, which remove the weight-related hyper-parameters and effectively avoid the vexing problem of hyper-parameters in multi-view graph clustering. Finally, experiments were carried out on synthetic data and different sizes benchmark data respectively to verify the validity of our method.

Effectiveness and Efficiency: Label-Aware Hierarchical Subgraph Learning for Protein–Protein Interaction

The study of protein–protein interactions (PPIs) holds immense significance in understanding various biological activities, as well as in drug discovery and disease diagnosis. Existing deep learning methods for PPI prediction, including graph neural networks (GNNs), have been widely employed as the solutions, while they often experience a decline in performance in the real world. We claim that the topological shortcut is one of the key problems contributing negatively to the performance, according to our analysis. By modeling the PPIs as a graph with protein as nodes and interactions as edge types, the prevailing models tend to learn the pattern of nodes’ degrees rather than intrinsic sequence-structure profiles, leading to the problem termed topological shortcut. The huge data growth of PPI leads to intensive computational costs and challenges computing devices, causing infeasibility in practice. To address the discussed problems, we propose a label-aware hierarchical subgraph learning method (laruGL-PPI) that can effectively infer PPIs while being interpretable. Specifically, we introduced edge-based subgraph sampling to effectively alleviate the problems of topological shortcuts and high computing costs. Besides, the inner-outer connections of PPIs are modeled as a hierarchical graph, together with the dependencies between interaction types constructed by a label graph. Extensive experiments conducted across various scales of PPI datasets have conclusively demonstrated that the laruGL-PPI method surpasses the most advanced PPI prediction techniques currently available, particularly in the testing of unseen proteins. Also, our model can recognize crucial sites of proteins, such as surface sites for binding and active sites for catalysis.

GSearch: ultra-fast and scalable genome search by combining K-mer hashing with hierarchical navigable small world graphs.

Genome search and/or classification typically involves finding the best-match database (reference) genomes and has become increasingly challenging due to the growing number of available database genomes and the fact that traditional methods do not scale well with large databases. By combining k-mer hashing-based probabilistic data structures (i.e. ProbMinHash, SuperMinHash, Densified MinHash and SetSketch) to estimate genomic distance, with a graph based nearest neighbor search algorithm (Hierarchical Navigable Small World Graphs, or HNSW), we created a new data structure and developed an associated computer program, GSearch, that is orders of magnitude faster than alternative tools while maintaining high accuracy and low memory usage. For example, GSearch can search 8000 query genomes against all available microbial or viral genomes for their best matches (n= ∼318 000 or ∼3 000 000, respectively) within a few minutes on a personal laptop, using ∼6 GB of memory (2.5 GB via SetSketch). Notably, GSearch has an O(log(N)) time complexity and will scale well with billions of genomes based on a database splitting strategy. Further, GSearch implements a three-step search strategy depending on the degree of novelty of the query genomes to maximize specificity and sensitivity. Therefore, GSearch solves a major bottleneck of microbiome studies that require genome search and/or classification.

Multidimensional Representation of Semantic Relations between Physical Theories, Fundamental Constants and Units of Measurement with Formal Concept Analysis

We propose several hierarchical graphs that represent the semantic relations between physical theories, their fundamental constants and units of measurement. We begin with an alternative representation of Zel’manov’s cube of fundamental constants as a concept lattice. We then propose the inclusion of a new fundamental constant, Milgrom’s critical acceleration, and discuss the implications of such analysis. We then look for the same fundamental constants in a graph that relates magnitudes and units of measurement in the International System of Units. This exercise shows the potential of visualizing hierarchical networks as a tool to better comprehend the symmetries, interrelations and dependencies of physical magnitudes, units and theories. New regimes of application may be deduced, as well as an interesting reflection on our ontologies and corresponding theoretical objects.

Micro-expression recognition based on a novel GCN-transformer cooperation model for IoT-eHealth

If a person is truly healthy, his/her well-being encompasses both physical and psychological health. However, the existing IoT-eHealth system typically focus only on monitoring the user’s physical data through various sensors, neglecting their mental state. To enhance the intelligence level of IoT-eHealth system and enable it to have the psychological monitoring ability, a novel collaborative model based on Graph Convolutional Network (GCN) and Transformer is designed for Micro-Expression (ME) recognition in this paper. Firstly, facial information within each frame is transformed into a Spatial Topological Relationship Graph (STRG) by using facial landmarks detection and psychological relationship of local patches. Then, in order to automatically aggregate the key information on facial patches that contribute to ME recognition from the structured graph data, a Hierarchical Adaptive Graph Pooling (HAGP) module is designed for obtaining discriminative frame-level feature based on GCN utilizing graph structure and vertex global dependencies. Finally, in order to model the long-term dependencies among frames and capture the key frame-level features that are beneficial for ME recognition, a Temporal Sensitive Self-Attention (TSSA) mechanism is designed, and a novel Temporal Sensitive Transformer (TST) encoder is constructed based on TSSA to explore the evolution law of facial patterns and obtain discriminative video-level features that are helpful for ME recognition. In the comparative experiments of standard dataset verification and practical dataset testing, designed collaborative model is superior to other methods and can achieve the highest recognition accuracy, which almost can meet the application requirements of IoT-eHealth system.

A privacy-preserving federated graph learning framework for threat detection in IoT trigger-action programming

Trigger-Action Programming (TAP) is a common user-programming paradigm in Internet of Things (IoT) smart home platforms, allowing users to create customized automation rules to match IoT devices and network services. However, the potential security threats associated with TAP rules are often overlooked or underestimated by users. To address this issue, we propose PFTAP, a novel federated graph learning framework for threat detection of TAP rules while simultaneously protecting user data and privacy. First, we propose a hierarchical graph attention network. This network comprises intra-rule attention and inter-rule attention modules, which enable the learning of comprehensive feature representations for triggers and actions. By capturing the intricate relationships between different rules, the network enhances the detection accuracy of risky TAP rules. Moreover, our framework is based on federated learning and integrates symmetric encryption and local differential privacy techniques, aiming to safeguard user privacy from unauthorized access or tampering. To evaluate the effectiveness of our framework, we conduct experiments using an extensive dataset of IFTTT rules. The experimental results convincingly demonstrate that PFTAP outperforms state-of-the-art methods in terms of threat detection performance.

Hierarchical Bidirected Graph Convolutions for Large-Scale 3-D Point Cloud Place Recognition.

In this article, we present a novel hierarchical bidirected graph convolution network (HiBi-GCN) for large-scale 3-D point cloud place recognition. Unlike place recognition methods based on 2-D images, those based on 3-D point cloud data are typically robust to substantial changes in real-world environments. However, these methods have difficulty in defining convolution for point cloud data to extract informative features. To solve this problem, we propose a new hierarchical kernel defined as a hierarchical graph structure through unsupervised clustering from the data. In particular, we pool hierarchical graphs from the fine to coarse direction using pooling edges and fuse the pooled graphs from the coarse to fine direction using fusing edges. The proposed method can, thus, learn representative features hierarchically and probabilistically; moreover, it can extract discriminative and informative global descriptors for place recognition. Experimental results demonstrate that the proposed hierarchical graph structure is more suitable for point clouds to represent real-world 3-D scenes.

A Hierarchical Graph Neural Network Framework for Predicting Protein-Protein Interaction Modulators With Functional Group Information and Hypergraph Structure.

Accurate prediction of small molecule modulators targeting protein-protein interactions (PPIMs) remains a significant challenge in drug discovery. Existing machine learning-based models rely on manual feature engineering, which is tedious and task-specific. Recently, deep learning models based on graph neural networks have made remarkable progress in molecular representation learning. However, many graph-based approaches ignore molecular hierarchical structure modeling guided by domain knowledge. In chemistry, the functional groups of a molecule determine its interaction with specific targets. Therefore, we propose a hierarchical graph neural network framework (called HiGPPIM) for predicting PPIMs by integrating atom-level and functional group-level features of molecules. HiGPPIM constructs atom-level and functional group-level graphs based on chemical knowledge and learns graph representations using graph attention networks. Furthermore, a hypergraph attention network is designed in HiGPPIM to aggregate and transform two-level graph information. We evaluate the performance of HiGPPIM on eight PPI families and two prediction tasks, namely PPIM identification and potency prediction. Experimental results demonstrate that HiGPPIM achieves state-of-the-art performance on both tasks and that using functional group information to guide PPIM prediction is effective.

Mesh Neural Networks Based on Dual Graph Pyramids.

Deep neural networks (DNNs) have been widely used for mesh processing in recent years. However, current DNNs can not process arbitrary meshes efficiently. On the one hand, most DNNs expect 2-manifold, watertight meshes, but many meshes, whether manually designed or automatically generated, may have gaps, non-manifold geometry, or other defects. On the other hand, the irregular structure of meshes also brings challenges to building hierarchical structures and aggregating local geometric information, which is critical to conduct DNNs. In this paper, we present DGNet, an efficient, effective and generic deep neural mesh processing network based on dual graph pyramids; it can handle arbitrary meshes. First, we construct dual graph pyramids for meshes to guide feature propagation between hierarchical levels for both downsampling and upsampling. Second, we propose a novel convolution to aggregate local features on the proposed hierarchical graphs. By utilizing both geodesic neighbors and euclidean neighbors, the network enables feature aggregation both within local surface patches and between isolated mesh components. Experimental results demonstrate that DGNet can be applied to both shape analysis and large-scale scene understanding. Furthermore, it achieves superior performance on various benchmarks, including ShapeNetCore, HumanBody, ScanNet and Matterport3D. Code and models will be available at https://github.com/li-xl/DGNet.

Properties of the Laplacian spectra of certain basic and hierarchical graphs

Properties of the Laplacian spectra of certain basic and hierarchical graphs

Influential simplices mining via simplicial convolutional networks

The identification of influential simplices is crucial for understanding higher-order network dynamics. Yet, despite relatively mature research on influential nodes (0-simplices) mining, characterizing simplices’ influence and identifying influential simplices remain challenging due to notable discrepancies in vital nodes and vital simplices mining. In this paper, we propose a higher-order graph learning model, named influential simplices mining neural networks (ISMnet), to identify vital simplices in simplicial complexes. ISMnet leverages novel higher-order representations: hierarchical bipartite graphs and higher-order hierarchical (HoH) Laplacians, where target simplices are grouped into a hub set and can interact with other simplices. It also employs learnable graph convolution operators in each HoH Laplacian domain to capture interactions among simplices and can identify influential simplices of arbitrary order by changing the hub set. Notably, ISMnet addresses the limitations inherent in traditional graph neural networks that struggle with higher-order tasks, while seamlessly retaining the capability to exploit network topology and node features concurrently. Numerical results on empirical and synthetic datasets demonstrate that ISMnet significantly outperforms existing methods by at least 12% and 4%, respectively, in ranking 2-simplices. In general, this novel framework promises to serve as a potent tool in higher-order network analysis.

Document-Level Relation Extraction Based on Machine Reading Comprehension and Hybrid Pointer-sequence Labeling

Document-level relational extraction requires reading, memorization, and reasoning to discover relevant factual information in multiple sentences. It is difficult for the current hierarchical network and graph network methods to fully capture the structural information behind the document and make natural reasoning from the context. Different from the previous methods, this article reconstructs the relation extraction task into a machine reading comprehension task. Each pair of entities and relationships is characterized by a question template, and the extraction of entities and relationships is translated into identifying answers from the context. To enhance the context comprehension ability of the extraction model and achieve more precise extraction, we introduce large language models (LLMs) during question construction, enabling the generation of exemplary answers. Besides, to solve the multi-label and multi-entity problems in documents, we propose a new answer extraction model based on hybrid pointer-sequence labeling, which improves the reasoning ability of the model and realizes the extraction of zero or multiple answers in documents. Extensive experiments on three public datasets show that the proposed method is effective.

Document-level biomedical relation extraction via hierarchical tree graph and relation segmentation module.

Biomedical relation extraction at the document level (Bio-DocRE) involves extracting relation instances from biomedical texts that span multiple sentences, often containing various entity concepts such as genes, diseases, chemicals, variants, etc. Currently, this task is usually implemented based on graphs or transformers. However, most work directly models entity features to relation prediction, ignoring the effectiveness of entity pair information as an intermediate state for relation prediction. In this article, we decouple this task into a three-stage process to capture sufficient information for improving relation prediction. We propose an innovative framework HTGRS for Bio-DocRE, which constructs a hierarchical tree graph (HTG) to integrate key information sources in the document, achieving relation reasoning based on entity. In addition, inspired by the idea of semantic segmentation, we conceptualize the task as a table-filling problem and develop a relation segmentation (RS) module to enhance relation reasoning based on the entity pair. Extensive experiments on three datasets show that the proposed framework outperforms the state-of-the-art methods and achieves superior performance. Our source code is available at https://github.com/passengeryjy/HTGRS.

Contrastive learning based on hierarchical graph of microstructures through directed energy deposition process to establish process-structure–property relationship via autoencoder

n this study, an innovative algorithm is developed to transform microstructures to hierarchical graphs without manual feature engineering. The hierarchical graph comprises two layers. Initially, pixel data, which includes Euler angles, phase, and position, forms the pixel-wise graphs within individual grains. Following this foundation, the grains serve as nodes, constituting the second layer. Importantly, the hierarchical graph preserves essential measurement data and structural details for training machine learning models. After that, a contrastive learning model based on hierarchical graph is designed to capture representations of microstructures obtained by the electron backscatter diffraction (EBSD) technique. This model can be directly extended for other microscopy techniques, such as Scanning Electron Microscopy (SEM), Energy-Dispersive X-ray spectroscopy (EDX), and Transmission Electron Microscopy (TEM). Using the learned microstructure representations, an autoencoder model for the directed energy deposition (DED) is developed to establish the relationship among process parameters, microstructures, and material properties, completing the process-structure–property cycle quantitatively. The performance of the present model is naturally benchmarked against two other models: a contrastive learning model based on pixel-wise graph (without manual feature engineering) and a contrastive learning model based on grain-wise graph (employing manual feature engineering). The results of the present model highlight the potential in decoding the process-structure–property relationships in the DED process.

HSeq2Seq: Hierarchical graph neural network for accurate mobile traffic forecasting

Hierarchical graph neural networks (HGNNs) provide a feasible method for modeling complex spatiotemporal dependencies during mobile traffic forecasting. However, most existing studies have adopted spatial node clustering methods to construct a coarsened graph that overlooks the temporal correlation within the original station graph. Furthermore, existing state-of-the-art methods fail to fully exploit the cross-regional feature impacts on stations, which limits their ability to model nonlocal spatial dependency.To overcome these limitations, we proposed a hierarchical sequence-to-sequence (HSeq2Seq) approach that combines HGNNs with a sequence-to-sequence architecture (Seq2Seq) for mobile traffic forecasting. First, a spatiotemporal node clustering method was designed to construct the hierarchical structure. Second, a convolution encoder was employed to extract the local spatiotemporal features from a hierarchical perspective, and a novel attention-based feature fusion module was built to capture the nonlocal spatiotemporal features by identifying both the intra- and cross-regional feature impacts on the stations. Finally, a recurrent decoder is introduced to reweight the spatiotemporal features and recursively produce the final prediction.Extensive experiments on two real-world datasets demonstrate that our model outperforms state-of-the-art methods, while the ablation study also verifies the effectiveness of the spatiotemporal node clustering and the attention-based feature fusion module.

Entropy Enhanced Multiagent Coordination Based on Hierarchical Graph Learning for Continuous Action Space

Entropy Enhanced Multiagent Coordination Based on Hierarchical Graph Learning for Continuous Action Space

Affective EEG-based cross-session person identification using hierarchical graph embedding

Affective EEG-based cross-session person identification using hierarchical graph embedding

The biomedical knowledge graph of symptom phenotype in coronary artery plaque: machine learning-based analysis of real-world clinical data.

A knowledge graph can effectively showcase the essential characteristics of data and is increasingly emerging as a significant means of integrating information in the field of artificial intelligence. Coronary artery plaque represents a significant etiology of cardiovascular events, posing a diagnostic challenge for clinicians who are confronted with a multitude of nonspecific symptoms. To visualize the hierarchical relationship network graph of the molecular mechanisms underlying plaque properties and symptom phenotypes, patient symptomatology was extracted from electronic health record data from real-world clinical settings. Phenotypic networks were constructed utilizing clinical data and protein‒protein interaction networks. Machine learning techniques, including convolutional neural networks, Dijkstra's algorithm, and gene ontology semantic similarity, were employed to quantify clinical and biological features within the network. The resulting features were then utilized to train a K-nearest neighbor model, yielding 23 symptoms, 41 association rules, and 61 hub genes across the three types of plaques studied, achieving an area under the curve of 92.5%. Weighted correlation network analysis and pathway enrichment were subsequently utilized to identify lipid status-related genes and inflammation-associated pathways that could help explain the differences in plaque properties. To confirm the validity of the network graph model, we conducted coexpression analysis of the hub genes to evaluate their potential diagnostic value. Additionally, we investigated immune cell infiltration, examined the correlations between hub genes and immune cells, and validated the reliability of the identified biological pathways. By integrating clinical data and molecular network information, this biomedical knowledge graph model effectively elucidated the potential molecular mechanisms that collude symptoms, diseases, and molecules.