Dynamic network representation learning based on temporal neighborhood aggregation

The link prediction problem is a time-evolving model in network science that has simultaneously abetted myriad applications and experienced extensive methodological improvement. Inferring the possibility of emerging links in dynamic social networks, also known as the dynamic link prediction task, is complex and challenging. In contrast to the link prediction in cross-sectional networks, dynamic link prediction methods need to cater to the actor-level temporal changes and associated evolutionary information regarding their micro- (i.e., link formation/deletion) and mesoscale (i.e., community formation) network structure. With the advent of abundant community detection algorithms, the research community has examined community-aware link prediction strategies in static networks. However, the same task in dynamic networks where, apart from the actors and links among them, their community pattern is also dynamic, is yet to be explored. Evolutionary community-aware information, including the associated link structure and temporal neighborhood changes, can effectively be mined to build dynamic similarity metrics for dynamic link prediction. This study aims to develop and integrate such dynamic features with machine learning algorithms for link prediction tasks in dynamic social networks. It also compares the performances of these features against well-known similarity metrics (i.e., ResourceAllocation) for static networks and a time series-based link prediction strategy in dynamic networks. These proposed features achieved high-performance scores, representing them as prospective candidates for both dynamic link prediction tasks and modeling the network growth.

The goal of network representation learning is to embed nodes so as to encode the proximity structures of a graph into a continuous low-dimensional feature space. In this paper, we propose a novel algorithm called node2hash based on feature hashing for generating node embeddings. This approach follows the encoder-decoder framework. There are two main mapping functions in this framework. The first is an encoder to map each node into high-dimensional vectors. The second is a decoder to hash these vectors into a lower dimensional feature space. More specifically, we firstly derive a proximity measurement called expected distance as target which combines position distribution and co-occurrence statistics of nodes over random walks so as to build a proximity matrix, then introduce a set of T different hash functions into feature hashing to generate uniformly distributed vector representations of nodes from the proximity matrix. Compared with the existing state-of-the-art network representation learning approaches, node2hash shows a competitive performance on multi-class node classification and link prediction tasks on three real-world networks from various domains.

Dynamic network link prediction is a hot research problem in complex networks. Network representation learning can effectively learn the similarity of nodes and can be used for link prediction. Existing dynamic network representation learning methods mainly discrete window dynamic networks and then model them on static network snapshot graphs, which is difficult to effectively deal with dynamic networks with fine-grained temporal characteristics. In this paper, we propose a link prediction model that can learn complex temporal properties in dynamic networks. The model uses continuous time event sequences to represent dynamic networks, learns continuous temporal information and structural evolution features in the networks, and proposes a temporal attention-based information transfer mechanism to model the diffusion and aggregation of information in the networks. Finally, it transforms link prediction into a classification problem. The experiments are conducted on four real dynamic network datasets and simulated networks, using ap and auc as evaluation metrics. The experimental results of real networks demonstrate that the model can effectively learn the continuity of network evolution and obtain an effective node representation, thus improving the link prediction effect. The experimental results of the simulated network show that the effect of link prediction is related to the network model, but the model in this paper can still obtain better prediction results.

Dynamic network representation learning has recently attracted increasing attention because real-world networks evolve over time, that is nodes and edges join or leave the networks over time. Different from static networks, the representation learning of dynamic networks should not only consider how to capture the structural information of network snapshots, but also consider how to capture the temporal dynamic information of network structure evolution from the network snapshot sequence. From the existing work on dynamic network representation, there are two main problems: (1) A significant number of methods target dynamic networks, which only allow nodes to increase over time, not decrease, which reduces the applicability of such methods to real-world networks. (2) At present, most network-embedding methods, especially dynamic network representation learning approaches, use Euclidean embedding space. However, the network itself is geometrically non-Euclidean, which leads to geometric inconsistencies between the embedded space and the underlying space of the network, which can affect the performance of the model. In order to solve the above two problems, we propose a geometry-based dynamic network learning framework, namely DyLFG. Our proposed framework targets dynamic networks, which allow nodes and edges to join or exit the network over time. In order to extract the structural information of network snapshots, we designed a new hyperbolic geometry processing layer, which is different from the previous literature. In order to deal with the temporal dynamics of the network snapshot sequence, we propose a gated recurrent unit (GRU) module based on Ricci curvature, that is the RGRU. In the proposed framework, we used a temporal attention layer and the RGRU to evolve the neural network weight matrix to capture temporal dynamics in the network snapshot sequence. The experimental results showed that our model outperformed the baseline approaches on the baseline datasets.

Real-world networked systems often show dynamic properties with continuously evolving network nodes and topology over time. When learning from dynamic networks, it is beneficial to correlate all temporal networks to fully capture the similarity/relevance between nodes. Recent work for dynamic network representation learning typically trains each single network independently and imposes relevance regularization on the network learning at different time steps. Such a snapshot scheme fails to leverage topology similarity between temporal networks for progressive training. In addition to the static node relationships within each network, nodes could show similar variation patterns (e.g., change of local structures) within the temporal network sequence. Both static node structures and temporal variation patterns can be combined to better characterize node affinities for unified embedding learning. In this paper, we propose Graph Attention Evolving Networks (GAEN) for dynamic network embedding with preserved similarities between nodes derived from their temporal variation patterns. Instead of training graph attention weights for each network independently, we allow model weights to share and evolve across all temporal networks based on their respective topology discrepancies. Experiments and validations, on four real-world dynamic graphs, demonstrate that GAEN outperforms the state-of-the-art in both link prediction and node classification tasks.

Network representation learning is a key research field in network data mining. In this paper, we propose a novel multi-view network representation algorithm (MVNR), which embeds multi-scale relations of network vertices into the low dimensional representation space. In contrast to existing approaches, MVNR explicitly encodes higher order information using k-step networks. In addition, we introduce the matrix forest index as a kind of network feature, which can be applied to balance the representation weights of different network views. We also research the relevance amongst MVNR and several excellent research achievements, including DeepWalk, node2vec and GraRep and so forth. We conduct our experiment on several real-world citation datasets and demonstrate that MVNR outperforms some new approaches using neural matrix factorization. Specifically, we demonstrate the efficiency of MVNR on network classification, visualization and link prediction tasks.

Distributed computation in dynamic networks via random walks

<p><span>Due to network dynamics in self-organizing networks the resource discovery effort increases. To discover objects in unstructured peer-to-peer network, peers rely on traditional methods like flooding, random walk and probabilistic forwarding methods. With inadequate knowledge of paths, the peers have to flood the query message which creates incredible network traffic and overhead. Many of the previous works based on random walk were done in wired network. In this context random walk was better than flooding. But under MANETs random walk approach behaved differently increasing the overhead, due to frequent link failures incurred by mobility. Decentralized applications based on peer-to-peer computing are best candidates to run over such dynamic network. Issues of P2P service discovery in wired networks have been well addressed in several earlier works. This article evaluates the performance of random walk based resource discovery protocol over P2P Mobile Adhoc Network (MP2P) and suggests an improved scheme to suit MANET. Our version reduces the network overhead, lowers the battery power consumption, minimizes the query delay while providing equally good success rate. The protocol is validated through extensive NS-2 simulations. It is clear from the results that our proposed scheme is an alternative to the existing ones for such highly dynamic mobile network scenario.</span></p>

Objective: The investigation of brain functional network dynamics offers a promising approach to understanding network reorganization poststroke. This study aims to explore the dynamic network configurations associated with motor recovery in stroke patients and assess their predictive potential using multilayer network analysis. Methods: Resting-state functional magnetic resonance imaging data were collected from patients with subacute stroke within 2 weeks of onset and from matched healthy controls (HCs). Group-independent component analysis and a sliding window approach were utilized to construct dynamic functional networks. A multilayer network model was applied to quantify the switching rates of individual nodes, subnetworks, and the global network across the dynamic network. Correlation analyses assessed the relationship between switching rates and motor function recovery, while linear regression models evaluated the predictive potential of global network switching rate on motor recovery outcomes. Results: Stroke patients exhibited a significant increase in the switching rates of specific brain regions, including the medial frontal gyrus, precentral gyrus, inferior parietal lobule, anterior cingulate, superior frontal gyrus, and postcentral gyrus, compared to HCs. Additionally, elevated switching rates were observed in the frontoparietal network, default mode network, cerebellar network, and in the global network. These increased switching rates were positively correlated with baseline Fugl-Meyer assessment (FMA) scores and changes in FMA scores at 90 days poststroke. Importantly, the global network's switching rate emerged as a significant predictor of motor recovery in stroke patients. Conclusions: The reorganization of dynamic network configurations in stroke patients reveals crucial insights into the mechanisms of motor recovery. These findings suggest that metrics of dynamic network reorganization, particularly global network switching rate, may offer a robust predictor of motor recovery.

Decremental algorithm for adaptive routing incorporating traveler information

Network representation learning is receiving increasing attention from scholars. Among them, methods based on graph neural networks have become particularly popular. However, most existing methods currently only focus on networks with a single type of relations. In the real world, networks contain a wealth of diverse information, with multiple types of relationships between nodes. In this paper, we propose a graph neural network-based multiplex network representation learning model (GNMRL). We model nodes within each layer of the multiplex network by aggregating neighbor information. Additionally, since nodes are influenced by different network layers, we integrate node interaction information across network layers. We conducted systematic experiments on four datasets, and the results show that GNMRL outperforms other comparison methods in both link prediction and node classification tasks.

In this paper, we propose a new approach named D2NE, short for Deep Dynamic Network Embedding, to learn the vertex representations for dynamic networks. The algorithm utilizes the graph attention mechanism to refresh embeddings efficiently, in which each update associate with local information only. To address the missing data, which is a common phenomenon in real-world networks, we model the auxiliary side information to capture more information on vertex relations. D2NE is highly efficient to refresh embeddings for dynamic networks, even with standard real-world sparse networks. We conduct extensive experiments on several real-world dynamic networks to validate the performance of D2NE in the link prediction task. Both quantitative results and qualitative analysis verify the effectiveness and rationality of our D2NE method.

A multi-scale reconstruction method for the anomaly detection in stochastic dynamic networks

The growing heterogeneous data in the Internet effectively improves the performance of recommender systems (RS). The main problem faced by the traditional matrix factorization (MF) is how to fuse more heterogeneous information data in MF to improve the performance of RS. In view of this, this paper proposes an Extended Matrix Factorization (EMF) based on network representation learning. EMF integrates multiple types of data in Heterogeneous Information Network (HIN) to improve the accuracy of prediction rating. The nodes in HIN are first mapped to a low-dimensional representation vector by network representation learning (NRL). Then the representation vector is used as the input of the EMF, the parameters are optimized by the gradient descent, and finally the prediction model is obtained. The experiments on two real data sets show the effectiveness of the EMF. Compared with the baseline algorithms, the EMF model can obtain more accurate prediction rating.

BackgroundReverse-engineering approaches such as Bayesian network inference, ordinary differential equations (ODEs) and information theory are widely applied to deriving causal relationships among different elements such as genes, proteins, metabolites, neurons, brain areas and so on, based upon multi-dimensional spatial and temporal data. There are several well-established reverse-engineering approaches to explore causal relationships in a dynamic network, such as ordinary differential equations (ODE), Bayesian networks, information theory and Granger Causality.ResultsHere we focused on Granger causality both in the time and frequency domain and in local and global networks, and applied our approach to experimental data (genes and proteins). For a small gene network, Granger causality outperformed all the other three approaches mentioned above. A global protein network of 812 proteins was reconstructed, using a novel approach. The obtained results fitted well with known experimental findings and predicted many experimentally testable results. In addition to interactions in the time domain, interactions in the frequency domain were also recovered.ConclusionsThe results on the proteomic data and gene data confirm that Granger causality is a simple and accurate approach to recover the network structure. Our approach is general and can be easily applied to other types of temporal data.