Real-world networks feature weights of interactions, where link often represent some physical attributes. In many situations, to recover the missing data or predict network evolution, we need in a network. this paper, first proposed series new centrality indices for links line graph. Then, utilizing these graph indices, as well number original designed three supervised learning methods realize weight prediction both single layer and multiple layers, which perform much better than several...

Community detection plays an important role in social networks, since it can help to naturally divide the network into smaller parts so as simplify analysis. However, on other hand, arises concern that individual information may be overmined, and concept community deception has been proposed protect privacy networks. Here, we introduce formalize problem of attack develop efficient strategies algorithms by rewiring a small number connections, leading protection. In particular, first give two...

Sequential recommender systems aim to predict users' next interested item given their historical interactions. However, a long-standing issue is how distinguish between long/short-term interests, which may be heterogeneous and contribute differently the recommendation. Existing approaches usually set pre-defined short-term interest length by exhaustive search or empirical experience, either highly inefficient yields subpar results. The recent advanced transformer-based models can achieve...

Real-world networks exhibit prominent hierarchical and modular structures, with various subgraphs as building blocks. Most existing studies simply consider distinct motifs use only their numbers to characterize the underlying network. Although such statistics can be used describe a network model, or even design some algorithms, role of in applications further explored so improve results. In this article, concept subgraph (SGN) is introduced then applied models, algorithms designed for...

In business domains, bundling is one of the most important marketing strategies to conduct product promotions, which commonly used in online e-commerce and offline retailers. Existing recommender systems mostly focus on recommending individual items that users may be interested in. this paper, we target at a practical but less explored recommendation problem named bundle recommendation, aims offer combination users. To tackle specific context virtual mall games, formalize it as link...

In social networks, by removing some target-sensitive links, privacy protection might be achieved. However, hidden links can still re-observed link prediction methods on observable networks. this paper, the conventional method named Resource Allocation Index (RA) is adopted for attacks. Several defense are proposed, including heuristic and evolutionary approaches, to protect targeted from RA attack. particular, incremental computation proposed accelerating calculation of fitness in...

Recent years have witnessed the great accuracy performance of graph-based Collaborative Filtering (CF) models for recommender systems. By taking user-item interaction behavior as a graph, these CF borrow success Graph Neural Networks (GNN), and iteratively perform neighborhood aggregation to propagate collaborative signals. While conventional are known facing challenges popularity bias that favors popular items, one may wonder "Whether existing alleviate or exacerbate systems?" To answer...

Community detection, aiming to group nodes based on their connections, plays an important role in network analysis since communities, treated as meta-nodes, allow us create a large-scale map of simplify its analysis. However, for privacy reasons, we may want prevent communities from being discovered certain cases, leading the topics community deception. In this article, formalize detection attack problem three scales, including global (macroscale), target (mesoscale), and node (microscale)....

Self-attention models have achieved state-of-the-art performance in sequential recommender systems by capturing the dependencies among user-item interactions. However, they rely on positional embeddings to retain information, which may break semantics of item embeddings. In addition, most existing works assume that such exist solely embeddings, but neglect their existence features. this work, we propose a novel system (MLP4Rec) based recent advances MLP-based architectures, is naturally...

Target search for moving and invisible objects has always been considered a challenge, as the floating drift with flows. This study focuses on target by multiple autonomous underwater vehicles (AUV) investigates multi-agent method (MATSMI) objects. In MATSMI algorithm, based deep deterministic policy gradient (MADDPG) method, we add spatial temporal information to reinforcement learning state set up specialized rewards in conjunction maritime scenario. Additionally, construct simulation...

Skeleton-based action recognition methods are limited by the semantic extraction of spatio-temporal skeletal maps. However, current have difficulty in effectively combining features from both temporal and spatial graph dimensions tend to be thick on one side thin other. In this paper, we propose a Temporal-Channel Aggregation Graph Convolutional Networks (TCA-GCN) learn topologies dynamically efficiently aggregate topological different channel for skeleton-based recognition. We use Temporal...

The underwater environment is still unknown for humans, so the high definition camera an important tool data acquisition at short distances underwater. Due to insufficient power, image collected by submersible devices cannot be analyzed in real time. Based on characteristics of Field-Programmable Gate Array (FPGA), low power consumption, strong computing capability, and flexibility, we design embedded FPGA recognition system Convolutional Neural Network (CNN). By using two technologies FPGA,...

Reinforcement learning based recommender systems (RL-based RS) aim at a good policy from batch of collected data, by casting recommendations to multi-step decision-making tasks. However, current RL-based RS research commonly has large reality gap. In this paper, we introduce the first open-source real-world dataset, RL4RS, hoping replace artificial datasets and semi-simulated previous studies used due resource limitation domain. Unlike academic RL research, suffers difficulties being...

Skeleton-based action recognition relies on the extraction of spatial-temporal topological information. Hypergraphs can establish prior unnatural dependencies for skeleton. However, existing methods only focus construction spatial topology and ignore time-point dependence. This paper proposes a dynamic hypergraph convolutional network (DST-HCN) to capture information skeleton-based recognition. DST-HCN introduces (TPH) learn relationships at time points. With multiple static hypergraphs TPH,...

Self-attention models have achieved the state-of-the-art performance in sequential recommender systems by capturing dependencies among user–item interactions. However, they rely on adding positional embeddings to item sequence retain information, which may break semantics of due heterogeneity between these two types embeddings. In addition, most existing works assume that such exist solely embeddings, but neglect their existence features. our previous study, we proposed a novel...

Abstract Distributed in-memory databases are widely adopted to achieve low latency and high bandwidth for data-intensive applications. They support scale-out by sharding distributing data across multiple nodes. To efficiently adapt various workloads, distributed must be capable of migrating shards In this paper, we demonstrate that state-of-the-art approaches experience significant performance degradation during migration due service downtime redundant transfer. Furthermore, our findings...

In recent years, extreme weather events accompanying the global warming have occurred frequently, which brought significant impact on national economic and social development. The ocean is an important member of climate system plays role in occurrence anomalies. With continuous improvement sensor technology, we use sensors to acquire data for study resource detection disaster prevention, etc. However, acquired by not enough be used directly researchers, so Generative Adversarial Network...

It is always difficult to evacuate crowds in public places like subway stations. The traditional crowd behavior simulation models often ignore two important issues evacuation: pedestrian tracking and individual differences. To solve the problem, this paper combines social force model (SFM) with deep learning into a novel detection method. Firstly, several algorithms for were compared, best ones sparse dense determined. Next, positions real video acquired by selected algorithms, converted...

In recent years, advances in Graph Convolutional Networks (GCNs) have given new insights into the development of social recommendation. However, many existing GCN-based recommendation methods often directly apply GCN to capture user-item and user-user interactions, which probably two main limitations: (a) Due power-law property degree distribution, vanilla with static normalized adjacency matrix has limitations learning node representations, especially for long-tail nodes; (b) multi-typed...

Online games make up the largest segment of booming global game market in terms revenue as well players. Unlike that sell at one time for profit, online money from in-game purchases by a large number engaged Therefore, Customer Lifetime Value (CLTV) is particularly vital companies to improve marketing decisions and increase revenues. Nowadays, virtual worlds are becoming increasingly innovative, complex, diverse, CLTV massive players highly personalized. That is, different may have very...

In typical alteration extraction methods, e.g., band math and principal component analysis (PCA), the bands or combinations unitized to extract altered minerals are usually selected based on empirical models previous rules. This results in significant differences of mineral mapping even same area, thus greatly increasing uncertainty resource prediction. this paper, an intelligent approach was proposed which optimization algorithm, i.e., a genetic algorithm (GA), introduced into PCA; is...