Search

To address the challenge of performance portability and facilitate the implementation of electronic structure solvers, we developed the basic matrix library (BML) and Parallel, Rapid O(N), and Graph-based Recursive Electronic Structure Solver (PROGRESS) library. The BML implements linear algebra operations necessary for electronic structure kernels using a unified user interface for various matrix formats (dense and sparse) and architectures (CPUs and GPUs). Focusing on density functional theory and tight-binding models, PROGRESS implements several solvers for computing the single-particle density matrix and relies on BML. In this paper, we describe the general strategies used for these implementations on various computer architectures, using OpenMP target functionalities on GPUs, in conjunction with third-party libraries to handle performance critical numerical kernels. We demonstrate the portability of this approach and its performance in benchmark problems. [ABSTRACT FROM AUTHOR]

Contemporary graph learning algorithms are not well-suited for large molecules since they do not consider the hierarchical interactions among the atoms, which are essential to determining the molecular properties of macromolecules. In this work, we propose Multiresolution Graph Transformers (MGT), the first graph transformer architecture that can learn to represent large molecules at multiple scales. MGT can learn to produce representations for the atoms and group them into meaningful functional groups or repeating units. We also introduce Wavelet Positional Encoding (WavePE), a new positional encoding method that can guarantee localization in both spectral and spatial domains. Our proposed model achieves competitive results on three macromolecule datasets consisting of polymers, peptides, and protein-ligand complexes, along with one drug-like molecule dataset. Significantly, our model outperforms other state-of-the-art methods and achieves chemical accuracy in estimating molecular properties (e.g., highest occupied molecular orbital, lowest unoccupied molecular orbital, and their gap) calculated by Density Functional Theory in the polymers dataset. Furthermore, the visualizations, including clustering results on macromolecules and low-dimensional spaces of their representations, demonstrate the capability of our methodology in learning to represent long-range and hierarchical structures. Our PyTorch implementation is publicly available at https://github.com/HySonLab/Multires-Graph-Transformer. [ABSTRACT FROM AUTHOR]

• A deep learning model for drug-target interaction prediction based on a knowledge graph and an intent graph is proposed, • Drug-target interactions be modeled from a more granular perspective, • Using relation path-aware aggregation under the GNN-base paradigm to integrate more information through a larger neighborhood region. Knowledge graph intent graph attention mechanism Predicting drug-target interactions (DTIs) plays a crucial role in drug discovery and drug development. Considering the high cost and risk of biological experiments, developing computational approaches to explore the interactions between drugs and targets can effectively reduce the time and cost of drug development. Recently, many methods have made significant progress in predicting DTIs. However, existing approaches still suffer from the high sparsity of DTI datasets and the cold start problem. In this paper, we develop a new model to predict drug-target interactions via a knowledge graph and intent graph named DTKGIN. Our method can effectively capture biological environment information for targets and drugs by mining their associated relations in the knowledge graph and considering drug-target interactions at a fine-grained level in the intent graph. DTKGIN learns the representation of drugs and targets from the knowledge graph and the intent graph. Then the probabilities of interactions between drugs and targets are obtained through the inner product of the representation of drugs and targets. Experimental results show that our proposed method outperforms other state-of-the-art methods in 10-fold cross-validation, especially in cold-start experimental settings. Furthermore, the case studies demonstrate the effectiveness of DTKGIN in predicting potential drug-target interactions. The code is available on GitHub: https://github.com/Royluoyi123/DTKGIN. [ABSTRACT FROM AUTHOR]

• Multi-entity fusion and low interpretability of answers are urgent issues • We employ heterogeneous graphs for entity fusion of community question answering • The heterogeneous graph neural network is used for entity relation embedding • Meta-paths represent user intention and are pioneered for answer interpretability • Experiments on real datasets verify the interpretable question-answering performance Community question answering is a rising technology based on users' autonomous interactive behaviors, such as posting their issues, answering questions based on their experience, and commenting on existing questions. As a result of its use of natural language for communication and stimulation of user interest in information sharing, it has increasingly taken the place of other channels as the main way that people learn new things. Multi-type entity characteristics fusion and poor answer interpretability are the two major concerns that currently plague community answer prediction research. The Interpretable Answer Retrieval Method Based on Heterogeneous Network Embedding (IARHNE) is what we present in this work. It combines complex entity features and generates interpretable predicted answers. In order to incorporate the interactions of several kinds of individuals in answer social retrieval, we first build a heterogeneity graph. In order to acquire entity embeddings, we secondly use the heterogeneous graph neural network. We then adopt the vector distance to convert the entity matching problem in the heterogeneous information network into a homogeneous node similarity job. Finally, using entity correlation to predict answers, we provide a list of answers to the new query and interpret them using meta-paths. Comparative studies using three authentic datasets demonstrate the benefits of IARHNE for interpretative question-answering research. [ABSTRACT FROM AUTHOR]

In light of the exponential growth in information volume, the significance of graph data has intensified. Graph clustering plays a pivotal role in graph data processing by jointly modeling the graph structure and node attributes. Notably, the practical significance of multi-view graph clustering is heightened due to the presence of diverse relationships within real-world graph data. Nonetheless, prevailing graph clustering techniques, predominantly grounded in deep learning neural networks, face challenges in effectively handling multi-view graph data. These challenges include the incapability to concurrently explore the relationships between multiple view structures and node attributes, as well as difficulties in processing multi-view graph data with varying features. To tackle these issues, this research proposes a straightforward yet effective multi-view graph clustering approach known as SLMGC. This approach uses graph filtering to filter noise, reduces computational complexity by extracting samples based on node importance, enhances clustering representations through graph contrastive regularization, and achieves the final clustering outcomes using a self-training clustering algorithm. Notably, unlike neural network algorithms, this approach avoids the need for intricate parameter settings. Comprehensive experiments validate the supremacy of the SLMGC approach in multi-view graph clustering endeavors when contrasted with prevailing deep neural network techniques. [ABSTRACT FROM AUTHOR]

Collaborative crowdsensing is a team collaboration model that harnesses the intelligence of a large network of participants, primarily applied in areas such as intelligent computing, federated learning, and blockchain. Unlike traditional crowdsensing, user recruitment in collaborative crowdsensing not only considers the individual capabilities of users but also emphasizes their collaborative abilities. In this context, this paper takes a unique approach by modeling user interactions as a graph, transforming the recruitment challenge into a graph theory problem. The methodology employs an enhanced Prim algorithm to identify optimal team members by finding the maximum spanning tree within the user interaction graph. After the recruitment, the collaborative crowdsensing explored in this paper presents a challenge of unfair incentives due to users engaging in free-riding behavior. To address these challenges, the paper introduces the MR-SVIM mechanism. Initially, the process begins with a Gaussian mixture model predicting the quality of users' tasks, combined with historical reputation values to calculate their direct reputation. Subsequently, to assess users' significance within the team, aggregation functions and the improved PageRank algorithm are employed for local and global influence evaluation, respectively. Indirect reputation is determined based on users' importance and similarity with interacting peers. Considering the comprehensive reputation value derived from the combined assessment of direct and indirect reputations, and integrating the collaborative capabilities among users, we have formulated a feature function for contribution. This function is applied within an enhanced Shapley value method to assess the relative contributions of each user, achieving a more equitable distribution of earnings. Finally, experiments conducted on real datasets validate the fairness of this mechanism. [ABSTRACT FROM AUTHOR]

We study in this paper single-peakedness on arbitrary graphs. Given a collection of preferences (rankings of alternatives), we aim at determining a connected graph G on which the preferences are single-peaked, in the sense that all the preferences are traversals of G. Note that a collection of preferences is always single-peaked on the complete graph. We propose an Integer Linear Programming formulation (ILP) of the problem of minimizing the number of edges in G or the maximum degree of a vertex in G. We prove that both problems are NP-hard in the general case. However, we show that if the optimal number of edges is m − 1 (where m is the number of candidates) then any optimal extreme point solution of the continuous relaxation of the ILP is integer and thus the integrality constraints can be relaxed. This provides an alternative proof of the polynomial time complexity of recognizing single-peaked preferences on a tree. We prove the same result for the case of a path (an axis), providing here also an alternative proof of polynomiality of the recognition problem. Furthermore, we provide a polynomial time procedure to recognize single-peaked preferences on a pseudotree (a connected graph that contains at most one cycle). We also give some experimental results, both on real and synthetic datasets. [ABSTRACT FROM AUTHOR]

New discoveries and knowledge are summarized in thousands of published papers per year per scientific domain, making it incomprehensible for scientists to account for all available knowledge relevant for their studies. In this paper, we present ACCORDION (ACCelerating and Optimizing model RecommenDatIONs), a novel methodology and an expert system that retrieves and selects relevant knowledge from literature and databases to recommend models with correct structure and accurate behavior, enabling mechanistic explanations and predictions, and advancing understanding. ACCORDION introduces an approach that integrates knowledge retrieval, graph algorithms, clustering, simulation, and formal analysis. Here, we focus on biological systems, although the proposed methodology is applicable in other domains. We used ACCORDION in nine benchmark case studies and compared its performance with other previously published tools. We show that ACCORDION is: comprehensive, retrieving relevant knowledge from a range of literature sources through machine reading engines; very effective, reducing the error of the initial baseline model by more than 80%, recommending models that closely recapitulate desired behavior, and outperforming previously published tools; selective, recommending only the most relevant, contextspecific, and useful subset (15%-20%) of candidate knowledge in literature; diverse, accounting for several distinct criteria to recommend more than one solution, thus enabling alternative explanations or intervention directions. [ABSTRACT FROM AUTHOR]

Traffic state classification and relevance calculation at intersections are both difficult problems in traffic control. In this paper, we propose an intersection relevance model based on a temporal graph attention network, which can solve the above two problems at the same time. First, the intersection features and interaction time of the intersections are regarded as input quantities together with the initial labels of the traffic data. Then, they are inputted into the temporal graph attention (TGAT) model to obtain the classification accuracy of the target intersections in four states—free, stable, slow moving, and congested—and the obtained neighbouring intersection weights are used as the correlation between the intersections. Finally, it is validated by VISSIM simulation experiments. In terms of classification accuracy, the TGAT model has a higher classification accuracy than the three traditional classification models and can cope well with the uneven distribution of the number of samples. The information gain algorithm from the information entropy theory was used to derive the average delay as the most influential factor on intersection status. The correlation from the TGAT model positively correlates with traffic flow, making it interpretable. Using this correlation to control the division of subareas improves the road network's operational efficiency more than the traditional correlation model does. This demonstrates the effectiveness of the TGAT model's correlation. [ABSTRACT FROM AUTHOR]

The integration of graph embedding technology and collaborative filtering algorithms has shown promise in enhancing the performance of recommendation systems. However, existing integrated recommendation algorithms often suffer from feature bias and lack effectiveness in personalized user recommendation. For instance, users' historical interactions with a certain class of items may inaccurately lead to recommendations of all items within that class, resulting in feature bias. Moreover, accommodating changes in user interests over time poses a significant challenge. This study introduces a novel recommendation model, RCKFM, which addresses these shortcomings by leveraging the CoFM model, TransR graph embedding model, backdoor tuning of causal inference, KL divergence, and the factorization machine model. RCKFM focuses on improving graph embedding technology, adjusting feature bias in embedding models, and achieving personalized recommendations. Specifically, it employs the TransR graph embedding model to handle various relationship types effectively, mitigates feature bias using causal inference techniques, and predicts changes in user interests through KL divergence, thereby enhancing the accuracy of personalized recommendations. Experimental evaluations conducted on publicly available datasets, including "MovieLens-1M" and "Douban dataset" from Kaggle, demonstrate the superior performance of the RCKFM model. The results indicate a significant improvement of between 3.17% and 6.81% in key indicators such as precision, recall, normalized discount cumulative gain, and hit rate in the top-10 recommendation tasks. These findings underscore the efficacy and potential impact of the proposed RCKFM model in advancing recommendation systems. [ABSTRACT FROM AUTHOR]

The need for speeding up data analytics increases inevitably due to the need for extracting valuable information from social media, data generated by smart devices with sensors, patterns of people's communications over the web, items viewed and bought by global-scale customers, cloud applications, etc., all of which take part in the "Big Data." Such kind of interaction data is very well represented as sparse graphs to enable the graph analytics, which requires efficient underlying kernels. The breadth-first search (BFS)-based traversal is a commonly used kernel in graph algorithms such as the betweenness centrality algorithm for centrality analysis. In this work, we focus on parallel BFS operations and propose hypergraph-based combinatorial models that aim at reducing cache misses and hence exploiting data locality during the parallel BFS operations. Our models are based on finding new vertex visit orders so that locality in accessing the data associated with vertices is exploited. Experiments on graphs arising in a wide range of applications show that our proposed models achieve on average 9% performance improvement in the CPU-based Ligra data analytics framework. [ABSTRACT FROM AUTHOR]

To address the matching challenge between the High Resolution Imaging Science Experiment (HiRISE) Digital Elevation Model (DEM) and the Mars Orbiter Laser Altimeter (MOLA) DEM, we propose a terrain matching framework based on the combination of point cloud coarse alignment and fine alignment methods. Firstly, we achieved global coarse localization of the HiRISE DEM through nearest neighbor matching of key Intrinsic Shape Signatures (ISS) points in the Fast Point Feature Histograms (FPFH) feature space. We introduced a graph matching strategy to mitigate gross errors in feature matching, employing a numerical method of non-cooperative game theory to solve the extremal optimization problem under Karush–Kuhn–Tucker (KKT) conditions. Secondly, to handle the substantial resolution disparities between the MOLA DEM and HiRISE DEM, we devised a smoothing weighting method tailored to enhance the Voxelized Generalized Iterative Closest Point (VGICP) approach for fine terrain registration. This involves leveraging the Euclidean distance between distributions to effectively weight loss and covariance, thereby reducing the results' sensitivity to voxel radius selection. Our experiments show that the proposed algorithm improves the accuracy of terrain registration on the proposed Curiosity landing area's, Mawrth Vallis, data by nearly 20%, with faster convergence and better algorithm robustness. [ABSTRACT FROM AUTHOR]

Image data augmentation plays a crucial role in data augmentation (DA) by increasing the quantity and diversity of labeled training data. However, existing methods have limitations. Notably, techniques like image manipulation, erasing, and mixing can distort images, compromising data quality. Accurate representation of objects without confusion is a challenge in methods like auto augment and feature augmentation. Preserving fine details and spatial relationships also proves difficult in certain techniques, as seen in deep generative models. To address these limitations, we propose OFIDA, an object-focused image data augmentation algorithm. OFIDA implements one-to-many enhancements that not only preserve essential target regions but also elevate the authenticity of simulating real-world settings and data distributions. Specifically, OFIDA utilizes a graph-based structure and object detection to streamline augmentation. Specifically, by leveraging graph properties like connectivity and hierarchy, it captures object essence and context for improved comprehension in real-world scenarios. Then, we introduce DynamicFocusNet, a novel object detection algorithm built on the graph framework. DynamicFocusNet merges dynamic graph convolutions and attention mechanisms to flexibly adjust receptive fields. Finally, the detected target images are extracted to facilitate one-to-many data augmentation. Experimental results validate the superiority of our OFIDA method over state-of-the-art methods across six benchmark datasets. [ABSTRACT FROM AUTHOR]

Algebraic multigrid (AMG) methods are often robust and effective solvers for solving the large and sparse linear systems that arise from discretized PDEs and other problems, relying on heuristic graph algorithms to achieve their performance. Reduction‐based AMG (AMGr) algorithms attempt to formalize these heuristics by providing two‐level convergence bounds that depend concretely on properties of the partitioning of the given matrix into its fine‐ and coarse‐grid degrees of freedom. MacLachlan and Saad (SISC 2007) proved that the AMGr method yields provably robust two‐level convergence for symmetric and positive‐definite matrices that are diagonally dominant, with a convergence factor bounded as a function of a coarsening parameter. However, when applying AMGr algorithms to matrices that are not diagonally dominant, not only do the convergence factor bounds not hold, but measured performance is notably degraded. Here, we present modifications to the classical AMGr algorithm that improve its performance on matrices that are not diagonally dominant, making use of strength of connection, sparse approximate inverse (SPAI) techniques, and interpolation truncation and rescaling, to improve robustness while maintaining control of the algorithmic costs. We present numerical results demonstrating the robustness of this approach for both classical isotropic diffusion problems and for non‐diagonally dominant systems coming from anisotropic diffusion. [ABSTRACT FROM AUTHOR]

• Oil retention models in various types of pipelines are established. • Hierarchical iterative algorithm based on graph theory is developed. • Oil retention in various pipelines is measured for model validation. • Mass flow rate or evaporation temperature increase may reduce oil retention. A multi-split air conditioning system may suffer damage caused by oil starvation in its compressor when the lubricating oil largely retains in long pipelines, and thus a reasonable oil charge based on accurate prediction of the oil retention in pipelines is important. The purpose of this study is to present an oil retention model for pipelines and to experimentally validate this model. The structure of pipelines connecting the outdoor unit and the indoor units with arbitrary numbers installed on separated floors is described as a combination of straight pipe types, joint types, and bend types. In modeling the oil retention, all the components of straight pipes, joints, and bends are numbered to conveniently distinguish each component which is further divided into several control volumes with equal lengths along the flow direction, and the equations of oil retention, pressure drop and enthalpy difference for the control volumes are established to describe the refrigerant-oil flow behavior. A hierarchical iterative algorithm based on the graph theory is developed in order to solve the models quickly. An experimental rig is built to measure the oil retention in four types of straight pipes, and the validation results show that the predicted oil retention in straight pipes agrees with at least 74.0 % of the experimental data within a deviation of ±20 %, and the mean deviation is lower than 15.4 %. Based on the model, the effects of oil retention reduction by increasing the total mass flow rate or the evaporation temperature of refrigerant are quantitatively analyzed. [ABSTRACT FROM AUTHOR]

Sparse arrays are widely used to achieve full array performance with fewer elements to reduce the cost of array and beamforming computation. Sparse array synthesis methods such as Bayesian compressed sensing (BCS) yield small element numbers; however, they are limited by the tradeoff between complexity and accuracy. Herein, a novel sparse array synthesis method with two-stage progressive BCS and an undirected graph-based element spacing constraint is proposed. The two-stage progressive BCS includes fast on-grid sparsification and accurate off-grid global re-estimation. First, the multitask BCS is solved using a relevance vector machine to efficiently select elements from candidate positions. Subsequently, a convex surrogate cost function is applied to the global re-estimation of the element weights to increase the beam pattern matching accuracy of the sparse array. Global optimization can improve the array performance. In addition, to satisfy the spacing constraint, a weighted merging method based on an undirected graph is proposed and inserted between the two stages to merge elements that are too close, which ensures the processability of the array. Simulations and experiments involving a variety of arrays were conducted to confirm the advantages of the method with regard to array sparsity, sidelobe suppression, beam pattern matching accuracy, and array processability. The proposed method achieved accurate and effective sparse array synthesis and outperformed existing methods. [ABSTRACT FROM AUTHOR]

Graph neural networks (GNNs), with their ability to incorporate node features into graph learning, have achieved impressive performance in many graph analysis tasks. However, current GNNs including the popular graph convolutional network (GCN) cannot obtain competitive results on the graphs without node features. In this work, we first introduce path-driven neighborhoods, and then define an extensional adjacency matrix as a convolutional operator. Second, we propose an approach named exopGCN which integrates the simple and effective convolutional operator into GCN to classify the nodes in the graphs without features. Experiments on six real-world graphs without node features indicate that exopGCN achieves better performance than other GNNs on node classification. Furthermore, by adding the simple convolutional operator into 13 GNNs, the accuracy of these methods are improved remarkably, which means that our research can offer a general skill to improve accuracy of GNNs. More importantly, we study the relationship between node classification by GCN without node features and community detection. Extensive experiments including six real-world graphs and nine synthetic graphs demonstrate that the positive relationship between them can provide a new direction on exploring the theories of GCNs. [ABSTRACT FROM AUTHOR]

As a knowledge representation method, knowledge graph is widely used in intelligent question answering systems and recommendation systems. At present, the research on knowledge graph mainly focuses on information query and retrieval based on knowledge graph. In some domain knowledge graphs, specific subgraph structures (patterns) have specific physical meanings. Aiming at this problem, this paper proposes a method and framework of knowledge graph pattern mining based on gat. Firstly, the patterns with specific physical meaning were transformed into subgraph structures containing topological structures and entity attributes. Secondly, the subgraph structure of the pattern is regarded as the query graph, and the knowledge graph is regarded as the data graph, so that the problem is transformed into an approximate subgraph matching problem. Then, the improved relational graph attention network is used to fuse the adaptive edge deletion mechanism to realize the approximate subgraph matching of subgraph structure and attribute, so as to obtain the best matching subgraph. The proposed method is trained in an end-to-end manner. The approximate subgraph matching is realized on the existing data set, and the research work of key pattern mining of complex geological structure knowledge graph is carried out. [ABSTRACT FROM AUTHOR]

The graph traversal edit distance (GTED), introduced by Ebrahimpour Boroojeny et al. (2018), is an elegant distance measure defined as the minimum edit distance between strings reconstructed from Eulerian trails in two edge-labeled graphs. GTED can be used to infer evolutionary relationships between species by comparing de Bruijn graphs directly without the computationally costly and error-prone process of genome assembly. Ebrahimpour Boroojeny et al. (2018) propose two ILP formulations for GTED and claim that GTED is polynomially solvable because the linear programming relaxation of one of the ILPs always yields optimal integer solutions. The claim that GTED is polynomially solvable is contradictory to the complexity results of existing string-to-graph matching problems. We resolve this conflict in complexity results by proving that GTED is NP-complete and showing that the ILPs proposed by Ebrahimpour Boroojeny et al. do not solve GTED but instead solve for a lower bound of GTED and are not solvable in polynomial time. In addition, we provide the first two, correct ILP formulations of GTED and evaluate their empirical efficiency. These results provide solid algorithmic foundations for comparing genome graphs and point to the direction of heuristics. The source code to reproduce experimental results is available at https://github.com/Kingsford-Group/gtednewilp/. [ABSTRACT FROM AUTHOR]

Knowledge graphs are now widely used in various domains, including Question-and-answer systems, intelligent search and recommendation systems, and intelligent decision-making systems. However, knowledge graphs inevitably contain inaccurate and incomplete knowledge during the creation process, which leads to a reduction in the usefulness of knowledge graphs. Therefore, to assess the usefulness of knowledge graphs based on specific application requirements, quality assessment is particularly important. Among them, accuracy assessment, as a necessary dimension, reflects the degree of correctness of the triples. However, in the actual assessment process, the existing assessment methods do not consider the user's needs and do not implement the concept of "Fitness for Use". Meanwhile, it takes a lot of labor cost to evaluate the accuracy of large-scale knowledge graphs. Therefore, to ensure the accuracy of the assessment in a cost-saving way while meeting the needs of users, we propose and implement a novel accuracy assessment method that focuses on the requirements of users by designing an effective sampling method to obtain accurate assessment results that are more practical and instructive for users. Finally, the performance of our proposed method is evaluated by comparing it with the real accuracy rate, and the experimental results show that the accuracy rate obtained by the proposed method is very close to the real accuracy rate, and the sample size is minimized. [ABSTRACT FROM AUTHOR]

Accurate atom mapping, which establishes correspondences between atoms in reactants and products, is a crucial step in analyzing chemical reactions. In this paper, we present a novel end-to-end approach that formulates the atom mapping problem as a deep graph matching task. Our proposed model, AMNet (Atom Matching Network), utilizes molecular graph representations and employs various atom and bond features using graph neural networks to capture the intricate structural characteristics of molecules, ensuring precise atom correspondence predictions. Notably, AMNet incorporates the consideration of molecule symmetry, enhancing accuracy while simultaneously reducing computational complexity. The integration of the Weisfeiler-Lehman isomorphism test for symmetry identification refines the model's predictions. Furthermore, our model maps the entire atom set in a chemical reaction, offering a comprehensive approach beyond focusing solely on the main molecules in reactions. We evaluated AMNet's performance on a subset of USPTO reaction datasets, addressing various tasks, including assessing the impact of molecular symmetry identification, understanding the influence of feature selection on AMNet performance, and comparing its performance with the state-of-the-art method. The result reveals an average accuracy of 97.3% on mapped atoms, with 99.7% of reactions correctly mapped when the correct mapped atom is within the top 10 predicted atoms. Scientific contribution The paper introduces a novel end-to-end deep graph matching model for atom mapping, utilizing molecular graph representations to capture structural characteristics effectively. It enhances accuracy by integrating symmetry detection through the Weisfeiler-Lehman test, reducing the number of possible mappings and improving efficiency. Unlike previous methods, it maps the entire reaction, not just main components, providing a comprehensive view. Additionally, by integrating efficient graph matching techniques, it reduces computational complexity, making atom mapping more feasible. [ABSTRACT FROM AUTHOR]

Graph domain adaptation (GDA) aims to address the challenge of limited label data in the target graph domain. Existing methods such as UDAGCN, GRADE, DEAL, and COCO for different-level (node-level, graph-level) adaptation tasks exhibit variations in domain feature extraction, and most of them solely rely on representation alignment to transfer label information from a labeled source domain to an unlabeled target domain. However, this approach can be influenced by irrelevant information and usually ignores the conditional shift of the downstream predictor. To effectively address this issue, we introduce a target-oriented unsupervised graph domain adaptive framework for graph adaptation called TO-UGDA. Particularly, domain-invariant feature representations are extracted using graph information bottleneck. The discrepancy between two domains is minimized using an adversarial alignment strategy to obtain a unified feature distribution. Additionally, the meta pseudo-label is introduced to enhance downstream adaptation and improve the model's generalizability. Through extensive experimentation on real-world graph datasets, it is proved that the proposed framework achieves excellent performance across various node-level and graph-level adaptation tasks. [ABSTRACT FROM AUTHOR]

The performance of semantic segmentation in remote sensing, based on deep learning models, depends on the training data. A commonly encountered issue is the imbalanced long-tailed distribution of data, where the head classes contain the majority of samples while the tail classes have fewer samples. When training with long-tailed data, the head classes dominate the training process, resulting in poorer performance in the tail classes. To address this issue, various strategies have been proposed, such as resampling, reweighting, and transfer learning. However, common resampling methods suffer from overfitting to the tail classes while underfitting the head classes, and reweighting methods are limited in the extreme imbalanced case. Additionally, transfer learning tends to transfer patterns learned from the head classes to the tail classes without rigorously validating its generalizability. These methods often lack additional information to assist in the recognition of tail class objects, thus limiting performance improvements and constraining generalization ability. To tackle the abovementioned issues, a graph neural network based on the graph kernel principle is proposed for the first time. By leveraging the graph kernel, structural information for tail class objects is obtained, serving as additional contextual information beyond basic visual features. This method partially compensates for the imbalance between tail and head class object information without compromising the recognition accuracy of head classes objects. The experimental results demonstrate that this study effectively addresses the poor recognition performance of small and rare targets, partially alleviates the issue of spectral confusion, and enhances the model's generalization ability. [ABSTRACT FROM AUTHOR]

Plant genomes include large numbers of transposable elements. One particular type of these elements is flanked by two Long Terminal Repeats (LTRs) and can translocate using RNA. Such elements are known as LTR-retrotransposons; they are the most abundant type of transposons in plant genomes. They have many important functions involving gene regulation and the rise of new genes and pseudo genes in response to severe stress. Additionally, LTR-retrotransposons have several applications in biotechnology. Due to the abundance and the importance of LTR-retrotransposons, multiple computational tools have been developed for their detection. However, none of these tools take advantages of the availability of related genomes; they process one chromosome at a time. Further, recently nested LTR-retrotransposons (multiple elements of the same family are inserted into each other) cannot be annotated accurately — or cannot be annotated at all — by the currently available tools. Motivated to overcome these two limitations, we built Look4LTRs, which can annotate LTR-retrotransposons in multiple related genomes simultaneously and discover recently nested elements. The methodology of Look4LTRs depends on techniques imported from the signal-processing field, graph algorithms, and machine learning with a minimal use of alignment algorithms. Four plant genomes were used in developing Look4LTRs and eight plant genomes for evaluating it in contrast to three related tools. Look4LTRs is the fastest while maintaining better or comparable F1 scores (the harmonic average of recall and precision) to those obtained by the other tools. Our results demonstrate the added benefit of annotating LTR-retrotransposons in multiple related genomes simultaneously and the ability to discover recently nested elements. Expert human manual examination of six elements — not included in the ground truth — revealed that three elements belong to known families and two elements are likely from new families. With respect to examining recently nested LTR-retrotransposons, three out of five were confirmed to be valid elements. Look4LTRs — with its speed, accuracy, and novel features — represents a true advancement in the annotation of LTR-retrotransposons, opening the door to many studies focused on understanding their functions in plants. [ABSTRACT FROM AUTHOR]

Efficient and accurate algorithms are necessary to reconstruct particles in the highly granular detectors anticipated at the High-Luminosity Large Hadron Collider and the Future Circular Collider. We study scalable machine learning models for event reconstruction in electron-positron collisions based on a full detector simulation. Particle-flow reconstruction can be formulated as a supervised learning task using tracks and calorimeter clusters. We compare a graph neural network and kernel-based transformer and demonstrate that we can avoid quadratic operations while achieving realistic reconstruction. We show that hyperparameter tuning significantly improves the performance of the models. The best graph neural network model shows improvement in the jet transverse momentum resolution by up to 50% compared to the rule-based algorithm. The resulting model is portable across Nvidia, AMD and Habana hardware. Accurate and fast machine-learning based reconstruction can significantly improve future measurements at colliders. The next generation of high energy particle colliders will have features that allows for highly granular detectors and current methods for particle collision reconstruction are limited. The authors explore machine learning algorithms for reconstructing events in electron-positron collisions for such future colliders obtaining a best-performing graph neural network that enhances the jet transverse momentum resolution by up to 50%, outperforming traditional methods and promising significant advancements in future collider measurements. [ABSTRACT FROM AUTHOR]

The influenza virus can spread easily, causing significant public health concern. Despite the existence of different techniques for rapid detection and prevention of influenza, their efficiency varies significantly. Additionally, there is currently a lack of a comprehensive, interoperable, and reusable real-time model for detecting influenza infection and predicting relationships within the field of influenza analysis. This study proposed a comprehensive, real-time model for rapid and early influenza detection using symptoms. Further, new relationships in the influenza field were discovered. Multiple data sources were used for the influenza knowledge graph (KG). Throughout this study, various graph algorithms were utilized to extract significant nodes and relationship features and multiple influenza detection machine learning (ML) models were compared. Node classification and link prediction methods were employed on a multi-layer perceptron (MLP) model. Furthermore, the hyperparameters of the model were automatically tuned. The proposed MLP model demonstrated the lowest rate of loss and the highest specificity, accuracy, recall, precision, and F1-score compared to state-of-the-art ML models. Moreover, the Matthews correlation coefficient was promising. This study shows that graph data science can improve MLP model detection and assist in discovering hidden connections in influenza KG. [ABSTRACT FROM AUTHOR]

Graph Neural Networks (GNNs) excel in compound property and activity prediction, but the choice of molecular graph representations significantly influences model learning and interpretation. While atom-level molecular graphs resemble natural topology, they overlook key substructures or functional groups and their interpretation partially aligns with chemical intuition. Recent research suggests alternative representations using reduced molecular graphs to integrate higher-level chemical information and leverages both representations for model. However, there is a lack of studies about applicability and impact of different molecular graphs on model learning and interpretation. Here, we introduce MMGX (Multiple Molecular Graph eXplainable discovery), investigating the effects of multiple molecular graphs, including Atom, Pharmacophore, JunctionTree, and FunctionalGroup, on model learning and interpretation with various perspectives. Our findings indicate that multiple graphs relatively improve model performance, but in varying degrees depending on datasets. Interpretation from multiple graphs in different views provides more comprehensive features and potential substructures consistent with background knowledge. These results help to understand model decisions and offer valuable insights for subsequent tasks. The concept of multiple molecular graph representations and diverse interpretation perspectives has broad applicability across tasks, architectures, and explanation techniques, enhancing model learning and interpretation for relevant applications in drug discovery. Reduced molecular graphs can integrate higher-level chemical information and leverage advantages from atom-level graph neural networks. Here, the authors introduce the Multiple Molecular Graph eXplainable model, investigating the effects of multiple molecular graphs, including Atom, Pharmacophore, JunctionTree, and FunctionalGroup, on model learning and interpretation from various perspectives [ABSTRACT FROM AUTHOR]

Motivation Proteoform identification is an important problem in proteomics. The main task is to find a modified protein that best fits the input spectrum. To overcome the combinatorial explosion of possible proteoforms, the proteoform mass graph and spectrum mass graph are used to represent the protein database and the spectrum, respectively. The problem becomes finding an optimal alignment between the proteoform mass graph and the spectrum mass graph. Peak error correction is an important issue for computing an optimal alignment between the two input mass graphs. Results We propose a faster algorithm for the error correction alignment of spectrum mass graph and proteoform mass graph problem and produce a program package TopMGFast. The newly designed algorithms require less space and running time so that we are able to compute global optimal alignments for the two input mass graphs in a reasonable time. For the local alignment version, experiments show that the running time of the new algorithm is reduced by 2.5 times. For the global alignment version, experiments show that the maximum mass errors between any pair of matched nodes in the alignments obtained by our method are within a small range as designed, while the alignments produced by the state-of-the-art method, TopMG, have very large maximum mass errors for many cases. The obtained alignment sizes are roughly the same for both TopMG and TopMGFast. Of course, TopMGFast needs more running time than TopMG. Therefore, our new algorithm can obtain more reliable global alignments within a reasonable time. This is the first time that global optimal error correction alignments can be obtained using real datasets. Availability and implementation The source code of the algorithm is available at https://github.com/Zeirdo/TopMGFast. [ABSTRACT FROM AUTHOR]

Traditional multi-label feature selection is usually performed under the condition of given label information, but nowadays, labeling multi-label data is a huge project, which is both time-consuming and labor-intensive, but if there is no label information condition, it will lead to poor feature selection, in order to solve this problem, this paper proposes a new semi-supervised multi-label feature selection method, i.e., semi-supervised multi-label feature selection algorithm based on dual dynamic graph. In this paper, a semi-supervised multi-label feature selection algorithm is proposed by constructing a dual dynamic graph. First, the method selects the most discriminative features for dimensionality reduction through the feature selection method of least squares regression, combined with the redundancy penalty of highly correlated features. Second, the label information is added to the construction of sample matrix similarity to learn the similarity. A semi-supervised multi-label feature selection framework is constructed by designing iterative updates of dual dynamic graphs to learn more accurate pseudo-label matrices to guide feature selection. Finally, the paper validates the above model using the alternating iteration optimization algorithm and verifies the effectiveness of the algorithm through experiments. [ABSTRACT FROM AUTHOR]

Social recommendations enhance the quality of recommendations by integrating social network information. Existing methods predominantly rely on pairwise relationships to uncover potential user preferences. However, they usually overlook the exploration of higher-order user relations. Moreover, because social relation graphs often exhibit scale-free graph structures, directly embedding them in Euclidean space will lead to significant distortion. To this end, we propose a novel graph neural network framework with hypergraph and hyperbolic embedding learning, namely HMGCN. Specifically, we first construct hypergraphs over user-item interactions and social networks, and then perform graph convolution on the hypergraphs. At the same time, a multi-channel setting is employed in the convolutional network, with each channel encoding its corresponding hypergraph to capture different high-order user relation patterns. In addition, we feed the item embeddings and the obtained high-order user embeddings into a hyperbolic graph convolutional network to extract user and item representations, enabling the model to better capture the hierarchical structure of their complex relationships. Experimental results on three public datasets, namely FilmTrust, LastFM, and Yelp, demonstrate that the model achieves more comprehensive user and item representations, more accurate fitting and processing of graph data, and effectively addresses the issues of insufficient user relationship extraction and data embedding distortion in social recommendation models. [ABSTRACT FROM AUTHOR]

In addition to providing learners with a large amount of teaching resources, online teaching platforms can also provide learning resources and channels such as video courseware, Q&A tutoring groups, and forums. However, currently, there are still shortcomings in depth and dimensionality in mining student learning behavior data on the platform. In view of this situation, based on the learning interaction behavior, this study established the difficulty similarity model of knowledge points, and used spectral clustering to classify their difficulty. In addition, the study intended to use the maximum frequent subgraph under the Gspan framework to characterize learners' implicit learning patterns. The outcomes expressed that the algorithm put forward in the study achieved the highest accuracy index of 98.8%, which was 1.4%, 4.0%, and 8.6% higher than Apriori-based graph mining algorithms, K-means, and frequent subgraph discovery algorithms. In terms of F1 index, the convergence value of the algorithm proposed in the study was 95.5%, which was about 2.5% higher than the last three algorithms. In addition, learners of all three cognitive levels had the highest maximum number of frequent subgraphs with sizes above 100 when the minSup value was 60%. And when the number of clusters was 3, the clustering accuracy of the three learners was the highest. In similarity calculation, the calculation method used in the study was at the minimum in terms of root mean square error and absolute error average index, which were 0.048% and 0.01% respectively. This indicated that the model proposed by the research had better classification effect on the difficulty of knowledge points for learners of different cognitive levels, and had certain application potential. [ABSTRACT FROM AUTHOR]

The canonical decomposition of a bipartite graph is a new decomposition method that involves three operators: parallel, series, and K ⨁   S . The class of weak-bisplit graphs is the class of totally decomposable graphs with respect to these operators, and the class of bicographs is the class of totally decomposable graphs with respect to parallel and series operators. We prove in this paper that the class of bipartite (P 6 , C 6) -free graphs is the class of bipartite graphs that are totally decomposable with respect to parallel and K ⨁ S operators. We present a linear time recognition algorithm for (P 6 , C 6) -free graphs that is symmetrical to the linear recognition algorithms of weak-bisplit graphs and s t a r 1,2 , 3 -free bipartite graphs. As a result of this algorithm, we present efficient solutions in this class of graphs for two optimization graph problems: the maximum balanced biclique problem and the maximum independent set problem. [ABSTRACT FROM AUTHOR]

Recent advances in knowledge graphs show great promise to link various data together to provide a semantic network. Place is an important part in the big picture of the knowledge graph since it serves as a powerful glue to link any data to its georeference. A key technical challenge in constructing knowledge graphs with location nodes as geographical references is the matching of place entities. Traditional methods typically rely on rule-based matching or machine-learning techniques to determine if two place names refer to the same location. However, these approaches are often limited in the feature selection of places for matching criteria, resulting in imbalanced consideration of spatial and semantic features. Deep feature-based methods such as deep learning methods show great promise for improved place data conflation. This paper introduces a Semantic-Spatial Aware Representation Learning Model (SSARLM) for Place Matching. SSARLM liberates the tedious manual feature extraction step inherent in traditional methods, enabling an end-to-end place entity matching pipeline. Furthermore, we introduce an embedding fusion module designed for the unified encoding of semantic and spatial information. In the experiment, we evaluate the approach to named places from Guangzhou and Shanghai cities in GeoNames, OpenStreetMap (OSM), and Baidu Map. The SSARLM is compared with several classical and commonly used binary classification machine learning models, and the state-of-the-art large language model, GPT-4. The results demonstrate the benefit of pre-trained models in data conflation of named places. [ABSTRACT FROM AUTHOR]

The Internet of Vehicles (IoV) autonomous driving technology based on deep learning has achieved great success. However, under the tunnel environment, the computer vision‐based IoV may fail due to low illumination. In order to handle this issue, this paper deploys an image enhancement module at the terminal of the IoV to alleviate the low illumination influence. The enhanced images can be submitted through IoT to the cloud server for further processing. The core algorithm of image enhancement is implemented by a dynamic graph embedded transformer network based on federated learning which can fully utilize the data resources of multiple devices in IoV and improve the generalization. Extensive comparative experiments are conducted on the publicly available dataset and the self‐built dataset which is collected under the tunnel environment. Compared with other deep models, all results confirm that the proposed graph embedded Transformer model can effectively enhance the detail information of the low‐light image, which can greatly improve the following tasks in IoV. [ABSTRACT FROM AUTHOR]

In the last two decades, we are witnessing a huge increase of valuable big data structured in the form of graphs or networks. To apply traditional machine learning and data analytic techniques to such data it is necessary to transform graphs into vector‐based representations that preserve the most essential structural properties of graphs. For this purpose, a large number of graph embedding methods have been proposed in the literature. Most of them produce general‐purpose embeddings suitable for a variety of applications such as node clustering, node classification, graph visualization and link prediction. In this article, we propose two novel graph embedding algorithms based on random walks that are specifically designed for the node classification problem. Random walk sampling strategies of the proposed algorithms have been designed to pay special attention to hubs–high‐degree nodes that have the most critical role for the overall connectedness in large‐scale graphs. The proposed methods are experimentally evaluated by analyzing the classification performance of three classification algorithms trained on embeddings of real‐world networks. The obtained results indicate that our methods considerably improve the predictive power of examined classifiers compared with currently the most popular random walk method for generating general‐purpose graph embeddings (node2vec). [ABSTRACT FROM AUTHOR]

Spatial clusters contain biases and artifacts, whether they are defined via statistical algorithms or via expert judgment. Graph‐based partitioning of spatial data and associated heuristics gained popularity due to their scalability but can define suboptimal regions due to algorithmic biases such as chaining. Despite the broad literature on deterministic regionalization methods, approaches that quantify regionalization probability are sparse. In this article, we propose a local method to quantify regionalization probabilities for regions defined via graph‐based cuts and expert‐defined regions. We conceptualize spatial regions as consisting of two types of spatial elements: core and swing. We define three distinct types of regionalization biases that occur in graph‐based methods and showcase the use of the proposed method to capture these types of biases. Additionally, we propose an efficient solution to the probabilistic graph‐based regionalization problem via performing optimal tree cuts along random spanning trees within an evidence accumulation framework. We perform statistical tests on synthetic data to assess resulting probability maps for varying distinctness of underlying regions and regionalization parameters. Lastly, we showcase the application of our method to define probabilistic ecoregions using climatic and remotely sensed vegetation indicators and apply our method to assign probabilities to the expert‐defined Bailey's ecoregions. [ABSTRACT FROM AUTHOR]

The purpose of constructing onboard observation mission queues is to improve the execution efficiency of onboard tasks and reduce energy consumption, representing a significant challenge in achieving efficient global military reconnaissance and target tracking. Existing research often focuses on the aspect of task scheduling, aiming at optimizing the efficiency of single-task execution, while neglecting the complex dependencies that might exist between multiple tasks and payloads. Moreover, traditional task scheduling schemes are no longer suitable for large-scale tasks. To effectively reduce the number of tasks within the network, we introduce a network aggregation graph model based on multiple satellites and tasks, and propose a task aggregation priority dynamic calculation algorithm based on graph computations. Subsequently, we present a dynamic merging-based method for multi-satellite, multi-task aggregation, a novel approach for constructing onboard mission queues that can dynamically optimize the task queue according to real-time task demands and resource status. Simulation experiments demonstrate that, compared to baseline algorithms, our proposed task aggregation method significantly reduces the task size by approximately 25% and effectively increases the utilization rate of onboard resources. [ABSTRACT FROM AUTHOR]

Generally, performing multiple scans is necessary to cover entire scanning area, and multiple point clouds are thus obtained. These point clouds need to be registered together, which is the focus of our work. However, influenced by the noise, point density variation, partial overlap and so on, the success rate of registration is still low. For this reason, a pairwise registration method is first proposed. It includes two main components: the triple local coordinate image (TriLCI) descriptor and a new 3D transformation estimation technique. The descriptor has high descriptiveness and strong robustness, by which more right correspondences can be achieved. The 3D transformation estimation technique is employed to calculate the rotation matrix and translation vector under the condition that most of correspondences are false. In this technique, a new geometric distance constraint is developed, and is then applied to eliminate incorrect correspondences. The robust estimation based on the median is introduced to calculate the rotation matrix and translation vector. For multi-view registration, the connected graph algorithm and shape growing based method are applied. A comparative study is presented. The experimental results well demonstrate that the proposed pairwise registration method has high success rate. The comparative study gives the merits and demerits of the two multi-view registration methods. [ABSTRACT FROM AUTHOR]

The identification of drug-target interactions (DTI) is a valuable step in the drug discovery and repositioning process. However, traditional laboratory experiments are time-consuming and expensive. Computational methods have streamlined research to determine DTIs. The application of deep learning methods has significantly improved the prediction performance for DTIs. Modern deep learning methods can leverage multiple sources of information, including sequence data that contains biological structural information, and interaction data. While useful, these methods cannot be effectively applied to each type of information individually (e.g., chemical structure and interaction network) and do not take into account the specificity of DTI data such as low- or zero-interaction biological entities. To overcome these limitations, we propose a method called MFA-DTI (Multi-feature Fusion Adopted framework for DTI). MFA-DTI consists of three modules: an interaction graph learning module that processes the interaction network to generate interaction vectors, a chemical structure learning module that extracts features from the chemical structure, and a fusion module that combines these features for the final prediction. To validate the performance of MFA-DTI, we conducted experiments on six public datasets under different settings. The results indicate that the proposed method is highly effective in various settings and outperforms state-of-the-art methods. • A novel DTI prediction model that explores both the graph and structural features of drugs and proteins is proposed. • A graph processing algorithm for biological networks is designed. • A module for generating sequence structure vectors is developed. • The results on 6 public datasets show it outperforms existing methods in all observed metrics. [ABSTRACT FROM AUTHOR]

Several concepts that model processes of spreading (of information, disease, objects, etc.) in graphs or networks have been studied. In many contexts, we assume that some vertices of a graph G are contaminated initially, before the process starts. By the q -forcing rule, a contaminated vertex having at most q uncontaminated neighbors enforces all the neighbors to become contaminated, while by the p -percolation rule, an uncontaminated vertex becomes contaminated if at least p of its neighbors are contaminated. If given a set S of initially contaminated vertices all vertices eventually become contaminated when continuously applying the q -forcing rule (respectively the p -percolation rule), S is called a q -forcing set (respectively, a p -percolating set) in G. In this paper, we consider sets S that are at the same time q -forcing sets and p -percolating sets, and call them (p , q) -spreading sets. Given positive integers p and q , the minimum cardinality of a (p , q) -spreading set in G is a (p , q) -spreading number, σ (p , q) (G) , of G. While q -forcing sets have been studied in a dozen of papers, the decision version of the corresponding graph invariant has not been considered earlier, and we fill the gap by proving its NP-completeness. This, in turn, enables us to prove the NP-completeness of the decision version of the (p , q) -spreading number in graphs for an arbitrary choice of p and q. Again, for every p ∈ N and q ∈ N ∪ { ∞ } , we find a linear-time algorithm for determining the (p , q) -spreading number of a tree, where in the case p ≥ 2 we apply Riedl's algorithm from [Largest and smallest minimal percolating sets in trees, Electron. J. Combin. 19 (2012) Paper 64] on p -percolation in trees. In addition, we present a lower and an upper bound on the (p , q) -spreading number of a tree and characterize extremal families of trees. In the case of square grids, we combine some results of Bollobás from [The Art of Mathematics: Coffee Time in Memphis. Cambridge Univ. Press, New York, 2006], and the AIM Minimum Rank-Special Graphs Work Group from [Zero forcing sets and the minimum rank of graphs, Linear algebra Appl. 428 2008 1628–1648], and new results on (2 , 1) -spreading and (4 , q) -spreading to obtain σ (p , q) (P m □ P n) for all (p , q) ∈ (N ∖ { 3 }) × (N ∪ { ∞ }) and all m , n ∈ N. [ABSTRACT FROM AUTHOR]

Chemical graph theory, a subfield of graph theory, is used to investigate chemical substances and their characteristics. Chemical graph analysis sheds light on the connection, symmetry, and reactivity of molecules. It supports chemical property prediction, research of molecular reactions, drug development, and understanding of molecular networks. A crucial part of computational chemistry is chemical graph theory, which helps researchers analyze and manipulate chemical structures using graph algorithms and mathematical models. Beryllonitrene , a compound of interest due to its potential applications in various fields, is examined through the lens of graph theory and mathematical modeling. The study involves the calculation and interpretation of topological indices and graph entropy measures, which provide valuable insights into the structural and energetic properties of Beryllonitrene's molecular graph. Logarithmic regression models are employed to establish correlations between these indices, entropy, and other relevant molecular attributes. The results contribute to a deeper understanding of Beryllonitrene's complex characteristics, facilitating its potential applications in diverse scientific and technological domains. In this study, degree-based topological indices TI are determined, as well as the entropy of graphs based on these TI . [ABSTRACT FROM AUTHOR]

Our research focuses on extracting exchanged views from dialogical documents through argument pair extraction (APE). The objective of this process is to facilitate comprehension of complex argumentative discourse by finding the related arguments. The APE comprises two stages: argument mining and argument matching. Researchers typically employ sequence labeling models for mining arguments and text matching models to calculate the relationships between them, thereby generating argument pairs. However, these approaches fail to capture long-distance contextual information and struggle to fully comprehend the complex structure of arguments. In our work, we propose the context-aware heterogeneous graph matching (HGMN) model for the APE task. First, we design a graph schema specifically tailored to argumentative texts, along with a heterogeneous graph attention network that effectively captures context information and structural information of arguments. Moreover, the text matching between arguments is converted into a graph matching paradigm and a multi-granularity graph matching model is proposed to handle the intricate relationships between arguments at various levels of granularity. In this way, the semantics of argument are modeled structurally and thus capture the complicated correlations between arguments. Extensive experiments are conducted to evaluate the HGMN model, including comparisons with existing methods and the GPT series of large language models (LLM). The results demonstrate that HGMN outperforms the state-of-the-art method. [ABSTRACT FROM AUTHOR]

Inductive Logic Programming (ILP) is a research field at the intersection between machine learning and logic programming, focusing on developing a formal framework for inductively learning relational descriptions in the form of logic programs from examples and background knowledge. As an emerging method of ILP, Meta-Interpretive Learning (MIL) leverages the specialization of a set of higher-order metarules to learn logic programs. In MIL, the input includes a set of examples, background knowledge, and a set of metarules, while the output is a logic program. MIL executes a depth-first traversal search, where its program search space expands polynomially with the number of predicates in the provided background knowledge and exponentially with the number of clauses in the program, sometimes even leading to search collapse. To address this challenge, this study introduces a strategy that employs the concept of reuse, specifically through the integration of auxiliary predicates, to reduce the number of clauses in programs and improve the learning efficiency. This approach focuses on the proactive identification and reuse of common program patterns. To operationalize this strategy, we introduce MILER, a novel method integrating a predicate generator, program learner, and program evaluator. MILER leverages frequent subgraph mining techniques to detect common patterns from a limited dataset of training samples, subsequently embedding these patterns as auxiliary predicates into the background knowledge. In our experiments involving two Visual Question Answering (VQA) tasks and one program synthesis task, we assessed MILER's approach to utilizing reusable program patterns as auxiliary predicates. The results indicate that, by incorporating these patterns, MILER identifies reusable program patterns, reduces program clauses, and directly decreases the likelihood of timeouts compared to traditional MIL. This leads to improved learning success rates by optimizing computational efforts. [ABSTRACT FROM AUTHOR]

Hyperspectral image has excellent spectral information and abundant spatial information, and its feature quality is one of the key factors affecting the classification performance. Extreme learning machine (ELM) is widely used in hyperspectral classification problems. However, traditional ELM is usually based on regular Euclidean data, ignoring the inherent structured information between hyperspectral pixels, resulting in poor robustness. In this paper, we propose a new supervised extreme learning machine framework, termed spatial-spectral graph convolutional ELM (SSG-ELM), based on using graph convolution to extend ELM into the non-Euclidean domain. The method inherits all the advantages from ELM, and consists of a random graph convolutional layer followed by a graph convolutional regression layer, enabling it to model complex intra class variations. Otherwise, a local spectral-spatial context integration and reshaping mechanism is incorporated into the hidden layer feature representation by using a context-aware bilateral filtering procedure. Experimental results show that the proposed algorithm obtains a competitive performance and outperforms other state-of-the-art ELM-based methods. [ABSTRACT FROM AUTHOR]

This article delves into the analysis and identification of influential individuals within social networks, a fundamental issue in the realm of social network science and complex network analysis. Within social networks, the impact of individuals on information propagation, shaping beliefs, and influencing the behavior of others is of paramount importance. Most of the current methods are based on the assumption that the entire graph is visible. However, this assumption does not hold for many real-world graphs. This study is conducted to extend current maximization methods, with link prediction techniques to pseudo-visibility graphs. The proposed method in this paper leverages graph models and influence algorithms to identify and select influential individuals within a social network. It begins by employing graph models to predict connections within the network and subsequently employs influence algorithms to opt for the most suitable individuals for exerting influence within the network. The primary advantage of this approach lies in its capability to enhance communication and influence within social networks with minimal observation. This method aids in a better understanding of the network's structure and the behaviors of its members, ultimately facilitating more effective information dissemination. In this study, we implemented five concepts for influence maximization in the context of the proposed method. Our results indicate that the algorithms perform best when considering lower diffusion coefficients. This suggests that focusing on lower diffusion coefficients allows the algorithm to demonstrate its superiority. Additionally, our findings reveal a notable consistency in results across various graph removal scenarios, highlighting a key advantage of our proposed approach on real-world graphs. [ABSTRACT FROM AUTHOR]

Faced with the challenges of intelligently managing the entire product lifecycle, traditional management techniques no longer meet the complex demands of modern enterprises. To address it, we employ an ontology modeling method based on a multidimensional view of product information for a knowledge graph, culminating in a comprehensive PLM knowledge system. To better serve PLM users, we propose an intelligent recommendation model based on context and user behavior for knowledge recommendations. This model seamlessly integrates the knowledge storage structure, user preference attributes, and the rich semantics of the knowledge graph. To improve the recommendation capability of the model, we design an improved multi-hop neighbor node sampling algorithm and use a graph convolutional neural network to quantify the target user preference attributes layer by layer. After a series of rigorous validation experiments, our method consistently outperformed competing models across key performance metrics, such as accuracy and recall, underscoring its efficacy and practicality in knowledge recommendation within PLM systems. [ABSTRACT FROM AUTHOR]

Multi-object tracking (MOT) is a thriving research field in computer vision. The tracklet-based MOT frameworks are frequently employed to generate long and stable trajectories in work scenes that involve long-term occlusion. However, most of these methods train tracklet feature encoders using complex loss functions, lacking an end-to-end paradigm guided by association results, which ultimately leads to limited MOT performance. To address this issue, a graph-based tracklet association framework that seamlessly integrates tracklet feature learning with tracklet association, thereby achieving tracklet association in an end-to-end manner. Specifically, we perform tracklet-based MOT in the graph domain and transform the tracklet association problem into an edge classification task. A message passing network (MPN) is used to update the tracklet features globally, which enhances the robustness of the tracklet features. Additionally, an attention-based feature update function is proposed to ensure the significance of current object. The effectiveness of the proposed framework is demonstrated using MOT17 and MOT20 benchmark datasets, and the experimental results show that the graph-based tracklet association network is a model-independent and plug-and-play component that could combine with different frame-based trackers to boost the MOT performance significantly. [ABSTRACT FROM AUTHOR]