Explore the vast range of titles available.

In recent years, online social networks (OSNs) have thoroughly penetrated people’s lives. Since information always flows along with various online relations in OSNs, analysing these relations becomes one of the most fundamental problems in various applications. Unfortunately, limited by the privacy concerns, data availability and the pow-law distributions of most OSNs, we can not observe enough relation links all the time, which is difficult to improve downstream tasks. To address this problem, many studies try to predict potential relations in online social networks via existing pair-wise links. However, when the observable pair-wise links are extremely sparse, most of them fail to learn the smoothness on the networks and make their proposed methods brittle. In light of this, we go beyond pair-wise relations and leverage hypergraphs to learn higher relations in the graphs. A hypergraph allows one hyperedge to connect multiple nodes, which is perfect to include more potential pair-wise links and can guarantee smooth node embeddings for better link prediction performance. In this paper, we aim at predicting potential links in sparsely observed networks. To achieve this goal, we first start from some blurry hyperedges and then proposed a novel hyperedge shrinking method to make the learned hyperedges more hierarchical. This method can learn hypergraph structure automatically from the given sparsely observed links and rely less on manual design. Following this, we further propose a novel hypergraph-based graph neural network to learn potential links in the graph. To address semantic fusion in the heterogeneous networks, we put forward multi-level reconstruction methods to preserve both specific semantics denoted by meta-paths, and high-level semantics denoted by hypergraphs. We compare our method with four state-of-the-art baselines. Extensive evaluations demonstrate that our method can achieve the best linking prediction results, especially when the networks are sparse.

The fifth-generation mobile networks (5G) will not only enhance mobile broadband services, but also enable connectivity for a massive number of Internet-of-Things devices, such as wireless sensors, meters or actuators. Thus, 5G is expected to achieve a 1000-fold or more increase in capacity over 4G. The use of the millimeter-wave (mmWave) spectrum is a key enabler to allowing 5G to achieve such enhancement in capacity. To fully utilize the mmWave spectrum, 5G is expected to adopt a heterogeneous network (HetNet) architecture, wherein mmWave small cells are overlaid onto a conventional macro-cellular network. In the mmWave-integrated HetNet, splitting of the control plane (CP) and user plane (UP) will allow continuous connectivity and increase the capacity of the mmWave small cells. mmWave communication can be used not only for access linking, but also for wireless backhaul linking, which will facilitate the installation of mmWave small cells. In this study, a proof-of-concept (PoC) was conducted to demonstrate the practicality of a prototype mmWave-integrated HetNet, using mmWave technologies for both backhaul and access.

LoRa is popular for internet of things applications as this communication technology offers both a long range and a low power consumption. However, LoRaWAN, the standard MAC protocol that uses LoRa as physical layer, has the bottleneck of a high downlink latency to achieve energy efficiency. To overcome this drawback we explore the use of wake-up radio combined with LoRa, and propose an adequate MAC protocol that takes profit of both these heterogeneous and complementary technologies. This protocol allows an opportunistic selection of a cluster head that forwards commands from the gateway to the nodes in the same cluster. Furthermore, to achieve self-sustainability, sensor nodes might include an energy harvesting sub-system, for instance to scavenge energy from the light, and their quality of service can be tuned, according to their available energy. To have an effective self-sustaining LoRa system, we propose a new energy manager that allows less fluctuations of the quality of service between days and nights. Latency and energy are modeled in a hybrid manner, i.e., leveraging microbenchmarks on real hardware platforms, to explore the influence of the energy harvesting conditions on the quality of service of this heterogeneous network. It is clearly demonstrated that the cooperation of nodes within a cluster drastically reduces the latency of LoRa base station commands, e.g., by almost 90% compared to traditional LoRa scheme for a 10 nodes cluster.

The Internet of Things (IoT) connects numerous sensor nodes and devices, resulting in an increase in the bandwidth and data rates. However, this has led to a surge in data-hungry applications, which consume significant energy at battery-limited IoT nodes, causing rapid battery drainage. As a result, it is imperative to find a reliable solution that reduces the power consumption. A power optimization model utilizing a modified genetic algorithm is proposed to manage power resources efficiently and reduce high power consumption. In this model, each access point computes the optimal power using the modified genetic algorithm until it meets the fitness criteria and assigns it to each cellular user. Additionally, a weight-based user-scheduling algorithm is proposed to enhance network efficiency. This algorithm considers both the distance and received signal strength indicator (RSSI) to select a user for a specific base station. Furthermore, it assigns appropriate weights for the distance, and the RSSI helps increase the spectral efficiency performance. In this paper, the user-scheduling algorithm was assigned equal weights and combined with the power optimization model to analyze the power consumption and spectral efficiency performance metrics. The results demonstrated that the weight-based user-scheduling algorithm performed better and was supported by the optimal allocation of weights using a modified genetic algorithm. The outcome proved that the optimal allocation of transmission power for users reduced the cellular users’ power consumption and improved the spectral efficiency.

Rule mining is an important and challenging task in data mining. Although many state-of-art algorithms have been proposed on dense data, they are not effectively adaptive for sparse data, such as sparse heterogeneous networks. Transfer learning improves the performance of algorithms in the target domain by transferring knowledge from a similar source domain, which provides a feasible and effective method to solve the above challenge. In this paper, we propose a transfer learning-based algorithm to mine rules on sparse data effectively, named TL-ERMSD. The algorithm is capable of detecting the knowledge of a common structure as well as the rules and logics between the source and target domains. Then, rule transfer is carried out by establishing the mapping mechanism between the two domains. We conducted experiments over the heterogeneous network datasets, including the source domain dataset FB15K and the target domain dataset Yago2Sample. The results demonstrate that the proposed TL-ERMSD for rule mining has a significant advantage over the existing algorithms.

Identifying new disease indications for existing drugs can help facilitate drug development and reduce development cost. The previous drug–disease association prediction methods focused on data about drugs and diseases from multiple sources. However, they did not deeply integrate the neighbor topological information of drug and disease nodes from various meta-path perspectives. We propose a prediction method called NAPred to encode and integrate meta-path-level neighbor topologies, multiple kinds of drug attributes, and drug-related and disease-related similarities and associations. The multiple kinds of similarities between drugs reflect the degrees of similarity between two drugs from different perspectives. Therefore, we constructed three drug–disease heterogeneous networks according to these drug similarities, respectively. A learning framework based on fully connected neural networks and a convolutional neural network with an attention mechanism is proposed to learn information of the neighbor nodes of a pair of drug and disease nodes. The multiple neighbor sets composed of different kinds of nodes were formed respectively based on meta-paths with different semantics and different scales. We established the attention mechanisms at the neighbor-scale level and at the neighbor topology level to learn enhanced neighbor feature representations and enhanced neighbor topological representations. A convolutional-autoencoder-based module is proposed to encode the attributes of the drug–disease pair in three heterogeneous networks. Extensive experimental results indicated that NAPred outperformed several state-of-the-art methods for drug–disease association prediction, and the improved recall rates demonstrated that NAPred was able to retrieve more actual drug–disease associations from the top-ranked candidates. Case studies on five drugs further demonstrated the ability of NAPred to identify potential drug-related disease candidates.

People are associated with multiple social networks to enjoy the exclusive services provided by each. Such users may be well established in some networks but relatively new to others. In order to find their potential connections in any newly entered network, their already existing rich interactions in the neworks where they are well-established can be utilized. We consider two such heterogeneous bibliographic networks where there are common users and propose a novel, event-based embedding algorithm called ABHENE (Alignment Based HEterogeneous Network Embedding) using CNN and transfer learning to construct a target network in a low-dimensional space based on its aligned counterpart. This procedure is repeated for various time slots and the target networks so obtained are fed into an LSTM framework to produce an embedded target network at a future point in time. Using this projected network, the future links among various nodes are predicted. We compare ABHENE with other embedding methods and, from such analysis, it has been found that ABHENE outperforms all its counterparts. Ours is the first work to consider network alignment and link prediction across heterogeneous aligned dynamic social networks. The performance of our link prediction method too surpasses those of other state-of-the-art algorithms. The results highlight the fact that enhanced performance can be achieved by the inclusion of heterogeneity and dynamism in the prediction of future links in partially aligned social networks.

Reliable prediction of drug–target interaction (DTI) is essential for accelerating drug discovery, yet remains hindered by data imbalance, limited interpretability, and neglect of protein dynamics. Here, we present GHCDTI , a heterogeneous graph neural framework designed to overcome these challenges through three synergistic innovations. First, cross-view contrastive learning with adaptive positive sampling improves generalization under extreme class imbalance (positive/negative ratio<1:100). Second, heterogeneous data fusion integrates molecular graphs, protein structure graphs, and bioactivity data via cross-graph attention, enabling interpretable residue-level insights. Third, multi-scale wavelet feature extraction captures both conserved and dynamic structural features by decomposing protein conformations into frequency components. GHCDTI achieves state-of-the-art performance on benchmark datasets (AUC: 0.966 ± 0.016; AUPR: 0.888 ± 0.018) and processes 1,512 proteins and 708 drugs in under two minutes, highlighting its potential for scalable virtual screening and drug repositioning. These results demonstrate GHCDTI’s ability to effectively identify novel drug–target pairs, providing a practical tool for accelerating drug discovery and improving biomedical knowledge integration.

Background Correctly identifying the driver genes that promote cell growth can significantly assist drug design, cancer diagnosis and treatment. The recent large-scale cancer genomics projects have revealed multi-omics data from thousands of cancer patients, which requires to design effective models to unlock the hidden knowledge within the valuable data and discover cancer drivers contributing to tumorigenesis. Results In this work, we propose a graph convolution network-based method called MRNGCN that integrates multiple gene relationship networks to identify cancer driver genes. First, we constructed three gene relationship networks, including the gene–gene, gene–outlying gene and gene–miRNA networks. Then, genes learnt feature presentations from the three networks through three sharing-parameter heterogeneous graph convolution network (HGCN) models with the self-attention mechanism. After that, these gene features pass a convolution layer to generate fused features. Finally, we utilized the fused features and the original feature to optimize the model by minimizing the node and link prediction losses. Meanwhile, we combined the fused features, the original features and the three features learned from every network through a logistic regression model to predict cancer driver genes. Conclusions We applied the MRNGCN to predict pan-cancer and cancer type-specific driver genes. Experimental results show that our model performs well in terms of the area under the ROC curve (AUC) and the area under the precision–recall curve (AUPRC) compared to state-of-the-art methods. Ablation experimental results show that our model successfully improved the cancer driver identification by integrating multiple gene relationship networks.

Green technology is a new term which is used to describe the energy efficient technologies. In the context of mobile communications industry, complying with the green technology strategy is a challenge. This is because of the tradeoff between the Quality of Service (QoS) provided and the total energy used in the transmission. Reducing the transmission energy may cause degradation in the QoS, more distinctively, in highly populated areas. This paper explores the possibility of achieving the green technology goal in planning and deployment of the HetNet mobile network with efficient network QoS. A decoupled two stage multi-objective genetic algorithm is developed to provide the network base station distribution that would satisfy both the network QoS and green network demands. In the first stage the algorithm estimates the base station parameters for more energy efficient HetNet deployment for optimum network coverage. The initial base station candidate locations are provided by a network operator in Kuala Lumpu, Malaysia. The second stage of the developed algorithm selects the number and location of RS associated with each base station optimized in the first stage to improve the network capacity. To optimize the network power, a novel arrival rate based HetNet total power consumption model is derived to investigate the parameters that affect the network power expenditure. Results show that a remarkable energy saving of about 40 % of the operator transmission power could be achieved with full network coverage. The addition of RS associated with each base station would greatly improve network capacity on the expense of its power expenditure. The relative RS to base station capacity plays major rule in reducing HetNet power expenditure.