Impact of Heterogeneity on Network Embedding

Drug repositioning (also called drug repurposing) is a strategy for identifying new therapeutic targets for existing drugs. This approach is of great importance in pharmacology as it is a faster and cheaper way to develop new medical treatments. In this paper, we present, to our knowledge, the first application of multiplex-heterogeneous network embedding to drug repositioning. Network embedding learns the vector representations of nodes, opening the whole machine learning toolbox for a wide variety of applications including link prediction, node labelling or clustering. So far, the application of network embedding for drug repositioning focused on heterogeneous networks. Our approach for drug repositioning is based on multiplex-heterogeneous network embedding. Such method allows the richness and complexity of multiplex and heterogeneous networks to be projected in the same vector space. In other words, multiplex-heterogeneous networks aggregate different multi-omics data in the same network representation. We validate the approach on a task of link prediction and on a case study for SARS-CoV2 drug repositioning. Experimental results show that our approach is highly robust and effective for finding new drug-target associations.KeywordsNetwork embeddingMultiplex-heterogeneous networkMulti-layer networkDrug repositioningGraph representation learningSARS-CoV2COVID-19

Currently, online social networks and social media have become increasingly popular showing an exponential growth. This fact have attracted increasing research interest and, in turn, facilitating the emergence of new interdisciplinary research directions, such as social network analysis. In this scenario, link prediction is one of the most important tasks since it deals with the problem of the existence of a future relation among members in a social network. Previous techniques for link prediction were based on structural (or topological) information. Nevertheless, structural information is not enough to achieve a good performance in the link prediction task on large-scale social networks. Thus, the use of additional information, such as interests or behaviors that nodes have into their communities, may improve the link prediction performance. In this paper, we analyze the viability of using a set of simple and non-expensive techniques that combine structural with community information for predicting the existence of future links in a large-scale online social network, such as Twitter. Twitter, a microblogging service, has emerged as a useful source of informative data shared by millions of users whose relationships require no reciprocation. Twitter network was chosen because it is not well understood, mainly due to the occurrence of directed and asymmetric links yet. Experiments show that our proposals can be used efficiently to improve unsupervised and supervised link prediction task in a directed and asymmetric large-scale network.

Semantic models are utilized to add context information to datasets and make data accessible and understandable in applications such as dataspaces. Since the creation of such models is a time-consuming task that has to be performed by a human expert, different approaches to automate or support this process exist. A recurring problem is the task of link prediction, i.e., the automatic prediction of links between nodes in a graph, in this case semantic models, usually based on machine learning techniques. While, in general, semantic models are trained and evaluated on large reference datasets, these conditions often do not match the domain-specific real-world applications wherein only a small amount of existing data is available (the cold-start problem). In this study, we evaluated the performance of link prediction algorithms when datasets of a smaller size were used for training (few-shot scenarios). Based on the reported performance evaluation, we first selected algorithms for link prediction and then evaluated the performance of the selected subset using multiple reduced datasets. The results showed that two of the three selected algorithms were suitable for the task of link prediction in few-shot scenarios.

Neuro-Symbolic Artificial Intelligence (AI) focuses on integrating symbolic and sub-symbolic systems to enhance the performance and explainability of predictive models. Symbolic and sub-symbolic approaches differ fundamentally in how they represent data and make use of data features to reach conclusions. Neuro-symbolic systems have recently received significant attention in the scientific community. However, despite efforts in neural-symbolic integration, symbolic processing can still be better exploited, mainly when these hybrid approaches are defined on top of knowledge graphs. This work is built on the statement that knowledge graphs can naturally represent the convergence between data and their contextual meaning (i.e., knowledge). We propose a hybrid system that resorts to symbolic reasoning, expressed as a deductive database, to augment the contextual meaning of entities in a knowledge graph, thus, improving the performance of link prediction implemented using knowledge graph embedding (KGE) models. An entity context is defined as the ego network of the entity in a knowledge graph. Given a link prediction task, the proposed approach deduces new RDF triples in the ego networks of the entities corresponding to the heads and tails of the prediction task on the knowledge graph (KG). Since knowledge graphs may be incomplete and sparse, the facts deduced by the symbolic system not only reduce sparsity but also make explicit meaningful relations among the entities that compose an entity ego network. As a proof of concept, our approach is applied over a KG for lung cancer to predict treatment effectiveness. The empirical results put the deduction power of deductive databases into perspective. They indicate that making explicit deduced relationships in the ego networks empowers all the studied KGE models to generate more accurate links.

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.

With the large amount of complex network data that is increasingly available on the Web, link prediction has become a popular data-mining research field. The focus of this paper is on a link-prediction task that can be formulated as a binary classification problem in complex networks. To solve this link-prediction problem, a sparse-classification algorithm called “Truncated Kernel Projection Machine” that is based on empirical-feature selection is proposed. The proposed algorithm is a novel way to achieve a realization of sparse empirical-feature-based learning that is different from those of the regularized kernel-projection machines. The algorithm is more appealing than those of the previous outstanding learning machines since it can be computed efficiently, and it is also implemented easily and stably during the link-prediction task. The algorithm is applied here for link-prediction tasks in different complex networks, and an investigation of several classification algorithms was performed for comparison. The experimental results show that the proposed algorithm outperformed the compared algorithms in several key indices with a smaller number of test errors and greater stability.

Deep models can be made scale-invariant when trained with multi-scale information. Images can be easily made multi-scale, given their grid-like structures. Extending this to generic graphs poses major challenges. For example, in link prediction tasks, inputs are represented as graphs consisting of nodes and edges. Currently, the state-of-the-art model for link prediction uses supervised heuristic learning, which learns graph structure features centered on two target nodes. It then learns graph neural networks to predict the existence of links based on graph structure features. Thus, the performance of link prediction models highly depends on graph structure features. In this work, we propose a novel node aggregation method that can transform the enclosing subgraph into different scales and preserve the relationship between two target nodes for link prediction. A theory for analyzing the information loss during the re-scaling procedure is also provided. Graphs in different scales can provide scale-invariant information, which enables graph neural networks to learn invariant features and improve link prediction performance. Our experimental results on 14 datasets from different areas demonstrate that our proposed method outperforms the state-of-the-art methods by employing multi-scale graphs without additional parameters.

Knowledge graphs (KGs) are widely used in many real-world applications, such as information retrieval, question answering system, and personal recommendation. However, most KGs are suffering from the incompleteness problem. To deal with the task of link prediction, previous knowledge graph embedding methods require numerous reference instances for each relation. It is worth noting that most relations in KGs have only a few reference instances available. Existing works for few-shot link prediction evaluate the authenticity of triplets from a single relation perspective. In this paper, we propose Hybrid Matching Network (HMNet) for few-shot link prediction, evaluating triplets from entity and relation two perspectives. At the entity-aware matching network, HMNet uses attentive inductive embedding layer to aggregate entity features and relation-aware topology, and then provides entity-aware score to implement first perspective evaluation. At the relation-aware matching network, HMNet integrates feature attention mechanism to implement relation perspective evaluation. Experiments on two public datasets indicate that HMNet achieves promising performance in few-shot link prediction.

Network embedding is a useful tool to map graph structures into vector spaces, which facilitates graph analysis tasks including node classification, graph visualization, similarity calculation etc. Existing network embedding methods calculate embedding vectors based on node series generated by random walks. These methods treat all the links equally during the random walk procedure, which leads to the missing of structural information that is key to the embedding performance. We therefore propose in this paper a novel random walk-based network embedding method called Self-Adjusting Random Walk (SARW). SARW utilizes a self-adjusting strategy that makes the walking biased towards the links that are more strongly connected in order to better capture the structural information. Further more, the strengths of links are updated using the embedding output as feedback. Through experiments we have verified that our method out performs state-of-the-art network embedding methods, in node classification tasks and link prediction tasks.

OLMNE+FT: Multiplex network embedding based on overlapping links

Network representation learning algorithms, which aim at automatically encoding graphs into low-dimensional vector representations with a variety of node similarity definitions, have a wide range of downstream applications. Most existing methods either have low accuracies in downstream tasks or a very limited application field, such as article classification in citation networks. In this paper, we propose a novel network representation method, named Link Prediction based Network Representation (LPNR), which generalizes the latest graph neural network and optimizes a carefully designed objective function that preserves linkage structures. LPNR can not only learn meaningful node representations that achieve competitive accuracy in node centrality measurement and community detection but also achieve high accuracy in the link prediction task. Experiments prove the effectiveness of LPNR on three real-world networks. With the mini-batch and fixed sampling strategy, LPNR can learn the embedding of large graphs in a few hours.

Users of online social networks voluntarily participate in different user groups or communities. Researches suggest the presence of strong local community structure in these social networks, i.e., users tend to meet other people via mutual friendship. Recently, different approaches have considered communities structure information for increasing the link prediction accuracy. Nevertheless, these approaches consider that users belong to just one community. In this paper, we propose three measures for the link prediction task which take into account all different communities that users belong to. We perform experiments for both unsupervised and supervised link prediction strategies. The evaluation method considers the links imbalance problem. Results show that our proposals outperform state-of-the-art unsupervised link prediction measures and help to improve the link prediction task approached as a supervised strategy.KeywordsLink predictionsocial networkscommunitiessocial network analysisgraph mining

Link prediction is a technology to find existing but unobserved links (static) or predict new links (dynamic) by studying the network topology. In real life, most networks change with time. Some recent works focus on the dynamic prediction of networks, but they do not solve the weight problem in link prediction well. At the same time, the methods based on matrix decomposition and node embedding used in most link prediction tasks have some problems, such as huge amount of calculation or can’t better represent the time evolution. In this paper, we extend the method of predicting the weight between nodes in dynamic networks and propose a generative adversarial model based on multi-layer graph convolution and recurrent neural network. The model consists of a generator and discriminator trained during an adversarial process. Multi-layer graph convolution used to approximate the high order similarity of network graph, so as to compensate the time evolution characteristic that RNN network can only deal with low order similarity. Using the process of confrontation training, the model can learn the representation of robust time evolution and predict with high accuracy. Experiments on several real data sets show that our model has good adaptability better than several baseline models.

Real-world graphs are dynamic, constantly evolving with new interactions, such as financial transactions in financial networks. Temporal Graph Neural Networks (TGNNs) have been developed to effectively capture the evolving patterns in dynamic graphs. While these models have demonstrated their superiority, being widely adopted in various important fields, their vulnerabilities against adversarial attacks remain largely unexplored. In this paper, we propose T-SPEAR, a simple and effective adversarial attack method for link prediction on continuous-time dynamic graphs, focusing on investigating the vulnerabilities of TGNNs. Specifically, before the training procedure of a victim model, which is a TGNN for link prediction, we inject edge perturbations to the data that are unnoticeable in terms of the four constraints we propose, and yet effective enough to cause malfunction of the victim model. Moreover, we propose a robust training approach T-SHIELD to mitigate the impact of adversarial attacks. By using edge filtering and enforcing temporal smoothness to node embeddings, we enhance the robustness of the victim model. Our experimental study shows that T-SPEAR significantly degrades the victim model's performance on link prediction tasks, and even more, our attacks are transferable to other TGNNs, which differ from the victim model assumed by the attacker. Moreover, we demonstrate that T-SHIELD effectively filters out adversarial edges and exhibits robustness against adversarial attacks, surpassing the link prediction performance of the naive TGNN by up to 11.2% under T-SPEAR. The code and datasets are available at https://github.com/wooner49/T-spear-shield

Knowledge graph embedding models are used to learn low‐dimensional representations of entities and relations in knowledge graphs. In this paper, we propose Multi‐RAttE, an attention‐based learning method for multiple relational knowledge graph embedding representation, which divides the information transfer in the knowledge graph into cross‐relational information transfer and relation‐specific information transfer, and divides the embedding of knowledge graph entities into structural embedding and multi‐relational embedding for joint learning. To objectively analyse the performance of the Multi‐RAttE model, we select two typical datasets and different representative baseline models for experimental evaluation on several tasks such as link prediction, multi‐relation prediction and node classification. The experimental results show that the Multi‐RAttE model improves 8% over the state‐of‐the‐art model Composition‐based Multi‐Relational Graph Convolutional Networks (CompGCN) in terms of Hits@1 metric on the link prediction task on FB15k‐237 dataset; on the multi‐relation prediction task, the accuracy improves by 1.8% and 3.7% in the auc metric and F1 metric, respectively. The experimental results have proved that the Multi‐RAttE model can effectively perform the representation of multiple relations.