Mining dense subgraphs in a bipartite graph is fundamental task analysis, with numerous applications community detection, fraud and e-commerce recommendation. Existing subgraph models, such as biclique, k -biplex, -bitruss, (α,β)-core, often face challenges due to their high computational complexity or limitations effectively capturing the density of graph. To overcome these issues, this paper, we propose new model for graphs, namely (α,β)-dense subgraph, designed capture structure inherent...

Stenotrophomonas maltophilia is a Gram-negative bacterial pathogen of increasing concern to human health. Most clinical isolates S. efficiently form biofilms on biotic and abiotic surfaces, making this bacterium resistant number antibiotic treatments therefore difficult eliminate. To date, very few studies have investigated the molecular regulatory mechanisms responsible for biofilm formation. Here we constructed random transposon insertion mutant library ATCC 13637 screened 14,028 clones. A...

Ocean wireless sensor networks (OWSNs) play an important role in marine environment monitoring, underwater target tracking, and defense. OWSNs not only monitor the surface information real time but also act as relay layer for to establish data communication between sensors ship-based base stations, land-based satellites. The destructive resistance of is closely related where they are located. Affected by dynamics seawater, location nodes extremely easy shift, resulting deterioration...

Convolutional neural networks (CNNs) provide a dramatically powerful class of models, but are subject to traditional convolution that can merely aggregate permutation-ordered and dimension-equal local inputs. It causes CNNs allowed only manage signals on Euclidean or grid-like domains (e.g., images), not ones non-Euclidean graph traffic networks). To eliminate this limitation, we develop local-aggregation function, sharable nonlinear operation, permutation-unordered dimension-unequal inputs...

The effectiveness of knowledge graph embedding (KGE) largely depends on the ability to model intrinsic relation patterns and mapping properties. However, existing approaches can only capture some them with insufficient modeling capacity. In this work, we propose a more powerful KGE framework named HousE, which involves novel parameterization based two kinds Householder transformations: (1) rotations achieve superior capacity patterns; (2) projections handle sophisticated Theoretically, HousE...

Cohesive subgraph mining on attributed graphs is a fundamental problem in graph data analysis. Existing cohesive algorithms do not consider the <i>fairness</i> of attributes subgraph. In this article, we, for first time, introduce fairness into widely-used clique model to mine fairness-aware subgraphs. particular, we propose three novel maximal models graphs, called weak fair clique, strong and relative respectively. To enumerate all cliques, develop an efficient backtracking algorithm...

Maximal biclique enumeration is a fundamental problem in bipartite graph data analysis. Existing methods mainly focus on non-attributed graphs and also ignore the fairness of attributes. In this paper, we introduce concept into model for first time study fairness-aware enumeration. Specifically, propose two models, called single-side fair bi-side respectively. To efficiently enumerate all bicliques, present non-trivial pruning techniques, α-β core colorful pruning, to reduce size without...

Cohesive sub graph mining on attributed graphs is a fundamental problem in data analysis. Existing cohesive algorithms do not consider the fairness of attributes subgraph. In this paper, we for first time introduce into widely-used clique model to mine fairness-aware subgraphs. particular, propose two novel maximal models graphs, called weak fair and strong respectively. To enumerate all cliques, develop an efficient backtracking algorithm WFCEnum equipped with colorful k-core based pruning...

The International Maritime Dangerous Goods Code (IMDG Code) is the most important regulation in international maritime transport chain of dangerous goods. Any ship carrying goods must be strictly observed. Integrating and correlating cumbersome knowledge IMDG simplifying query process are great significance to safe transportation storage As a new method representation management, graph has been successfully applied many industries. It can present complex relationship between domain correlate...

This paper studies the computational offloading of CNN inference in dynamic multi-access edge computing (MEC) networks. To address uncertainties communication time and computation resource availability, we propose a novel semantic compression method, autoencoder-based architecture (AECNN), for effective extraction partial offloading. In encoder, introduce feature module based on channel attention mechanism CNNs, to compress intermediate data by selecting most informative features. decoder,...

Densest subgraph search, aiming to identify a with maximum edge density, faces limitations as the density inadequately reflects biases towards given vertex set R. To address this, R -subgraph was introduced, refining doubled by penalizing vertices in but not , using degree penalty factor. This advancement leads Anchored Subgraph (ADS) search problem, which finds Š highest for Nonetheless, current algorithms ADS face significant inefficiencies handling large-scale graphs or sizable set....

A wide variety of deep neural network models for graph-structured data have been proposed to solve tasks like node/graph classification and link prediction. By effectively learning low-dimensional embeddings graph nodes, they shown state-of-the-art performance. However, most existing learn node by exploring flat information propagation across the edges within local neighborhood each node. We argue that incorporating hierarchical can capture inherently topological features many realistic...

Logical reasoning over incomplete knowledge graphs to answer complex logical queries is a challenging task. With the emergence of new entities and relations in constantly evolving KGs, inductive KGs has become crucial problem. However, previous PLMs-based methods struggle model structures queries, which limits their ability generalize within same structure. In this paper, we propose structure-modeled textual encoding framework for KGs. It encodes linearized query using pre-trained language...

In recent years, contrastive learning achieves impressive results on self-supervised visual representation learning, but there still lacks a rigorous understanding of its dynamics. this paper, we show that if cast objective equivalently into the feature space, then dynamics admits an interpretable form. Specifically, gradient descent corresponds to specific message passing scheme corresponding augmentation graph. Based perspective, theoretically characterize how gradually learns...

Embedding models have shown great power in knowledge graph completion (KGC) task. By learning structural constraints for each training triple, these methods implicitly memorize intrinsic relation rules to infer missing links. However, this paper points out that the multi-hop are hard be reliably memorized due inherent deficiencies of such implicit memorization strategy, making embedding underperform predicting links between distant entity pairs. To alleviate problem, we present Vertical...

Mining cohesive subgraphs in attributed graphs is an essential problem the domain of graph data analysis. The integration fairness considerations significantly fuels interest models and algorithms for mining fairness-aware subgraphs. Notably, relative fair clique emerges as a robust model, ensuring not only comprehensive attribute coverage but also greater flexibility distributing vertices. Motivated by strength this we first time pioneer investigation into identification maximum large-scale...