Link Prediction Problem Research Articles

A novel similarity measure of link prediction in multi‐layer social networks based on reliable paths

AbstractOnline social networks are an integral element of modern societies and significantly influence the formation and consolidation of social relationships. In fact, these networks are multi‐layered so that there may be multiple links between a user' on different social networks. In this article, the link prediction problem for the same user in a two‐layer social network is examined, where we consider Twitter and Foursquare networks. Here, information related to the two‐layer communication is used to predict links in the Foursquare network. Link prediction aims to discover spurious links or predict the emergence of future links from the current network structure. There are many algorithms for link prediction in unweighted networks, however only a few have been developed for weighted networks. Based on the extraction of topological features from the network structure and the use of reliable paths between users, we developed a novel similarity measure for link prediction. Reliable paths have been proposed to develop unweight local similarity measures to weighted measures. Using these measures, both the existence of links and their weight can be predicted. Empirical analysis shows that the proposed similarity measure achieves superior performance to existing approaches and can more accurately predict future relationships. In addition, the proposed method has better results compared to single‐layer networks. Experiments show that the proposed similarity measure has an advantage precision of 1.8% over the Katz and FriendLink measures.

Temporal link prediction in directed networks based on self-attention mechanism

The development of graph neural networks (GCN) makes it possible to learn structural features from evolving complex networks. Even though a wide range of realistic networks are directed ones, few existing works investigated the properties of directed and temporal networks. In this paper, we address the problem of temporal link prediction in directed networks and propose a deep learning model based on GCN and self-attention mechanism, namely TSAM. The proposed model adopts an autoencoder architecture, which utilizes graph attentional layers to capture the structural feature of neighborhood nodes, as well as a set of graph convolutional layers to capture motif features. A graph recurrent unit layer with self-attention is utilized to learn temporal variations in the snapshot sequence. We run comparative experiments on four realistic networks to validate the effectiveness of TSAM. Experimental results show that TSAM outperforms most benchmarks under two evaluation metrics.

Cold-Start Link Prediction via Weighted Symmetric Nonnegative Matrix Factorization with Graph Regularization

Link prediction has attracted wide attention among interdisciplinary researchers as an important issue in complex network. It aims to predict the missing links in current networks and new links that will appear in future networks. Despite the presence of missing links in the target network of link prediction studies, the network it processes remains macroscopically as a large connected graph. However, the complexity of the real world makes the complex networks abstracted from real systems often contain many isolated nodes. This phenomenon leads to existing link prediction methods not to efficiently implement the prediction of missing edges on isolated nodes. Therefore, the cold-start link prediction is favored as one of the most valuable subproblems of traditional link prediction. However, due to the loss of many links in the observation network, the topological information available for completing the link prediction task is extremely scarce. This presents a severe challenge for the study of cold-start link prediction. Therefore, how to mine and fuse more available non-topological information from observed network becomes the key point to solve the problem of cold-start link prediction. In this paper, we propose a framework for solving the cold-start link prediction problem, a joint-weighted symmetric nonnegative matrix factorization model fusing graph regularization information, based on low-rank approximation algorithms in the field of machine learning. First, the nonlinear features in high-dimensional space of node attributes are captured by the designed graph regularization term. Second, using a weighted matrix, we associate the attribute similarity and first order structure information of nodes and constrain each other. Finally, a unified framework for implementing cold-start link prediction is constructed by using a symmetric nonnegative matrix factorization model to integrate the multiple information extracted together. Extensive experimental validation on five real networks with attributes shows that the proposed model has very good predictive performance when predicting missing edges of isolated nodes.

Markov random geometric graph, MRGG: A growth model for temporal dynamic networks

We introduce Markov Random Geometric Graphs (MRGGs), a growth model for temporal dynamic networks. It is based on a Markovian latent space dynamic: consecutive latent points are sampled on the Euclidean Sphere using an unknown Markov kernel; and two nodes are connected with a probability depending on a unknown function of their latent geodesic distance. More precisely, at each stamp-time k we add a latent point Xk sampled by jumping from the previous one Xk−1 in a direction chosen uniformly Yk and with a length rk drawn from an unknown distribution called the latitude function. The connection probabilities between each pair of nodes are equal to the envelope function of the distance between these two latent points. We provide theoretical guarantees for the non-parametric estimation of the latitude and the envelope functions. We propose an efficient algorithm that achieves those non-parametric estimation tasks based on an ad-hoc Hierarchical Agglomerative Clustering approach. As a by product, we show how MRGGs can be used to detect dependence structure in growing graphs and to solve link prediction problems.

Link prediction in organizational social network based on e-mail communication

Last years one can observe a growing interest of researchers and practitioner in organizational social networks based on different forms of relations between employees. Such networks may refer to formal or informal relations between employees. The former of these relations may stem for example from the position of each person in an organization, and the latter may refer to relations between employees based for example on trust or confidence. Social network analysis tools provide several measures which allow to analyze different properties of such networks. The paper focuses on link prediction in organizational social network, where relations between employees are defined by intensity of digital communication (e-mail channel) between them. Different similarity measures have been considered by Authors. Basing on similarity scores calculated for all non-existing links in the network, new potential links with highest likelihood of their formation have been discovered. The results of the experiment (predicted links) have been compared with the results obtained by solving a community detection problem in the same network. They have been also compared with a real structure of the organization investigated in the experiment. The experiment confirmed that the results of solving a link prediction problem in the organizational network may support the process of detecting communities in the network, but it strongly depends on the similarity measure used.

Interlayer link prediction based on multiple network structural attributes

Information may flow across multiple social network platforms and spread simultaneously on them. These platforms constitute a multiplex network in which each social network platform functions as a subnetwork. One problem in information propagation analysis on the multiplex network is to reveal the corresponding relationship of accounts belonging to the same user across different subnetworks. It can be summarized as the interlayer link prediction problem in a multiplex network. To predict more unobserved interlayer links accurately, most of the current structure-based approaches adopt an iterative strategy. However, iterative manners suffer from a big challenge of high time consumption. In this paper, we propose an interlayer link prediction framework based on multiple structural attributes (MulAtt). It calculates the matching degree of unmatched nodes once by leveraging the information of closed triad, intralayer links, matched neighbors and intralayer links of neighbors simultaneously to guarantee accuracy while reducing time consumption. We also propose a fast possible matched closed triad counting algorithm to improve the efficiency of obtaining the clues from the closed triad. Experiments on four widely used real-world multiplex networks demonstrate that the MulAtt framework can achieve better performance than several existing network structure-based methods in a non-iterative way.

A Random Walk with Restart Model Based on Common Neighbors for Predicting the Clinical Drug Combinations on Coronary Heart Disease.

As the approaching of the clinical big data era, the prediction of whether drugs can be used in combination in clinical practice is a fundamental problem in the analysis of medical data. Compared with high-throughput screening, it is more cost-effective to treat this problem as a link prediction problem and predict by algorithms. Inspired by the rule of combined clinical medication, a new computational model is proposed. The drug-drug combination was predicted by combining the number of adjacent complete subgraphs shared by the two points with the restart random walk algorithm. The model is based on the semisupervised random walk algorithm, and the same neighborhood is used to improve the random walk with restart (CN-RWR). The algorithm can effectively improve the prediction performance and assign a score to any combination of drugs. To fairly compare the predictive performance of the improved model with that of the random walk with restart model (RWR), a cross-validation of the two models on the same drug data was performed. The AUROC of CN-RWR and RWR under the LOOCV validation framework is 0.9741 and 0.9586, respectively, and the improved model results are more reliable. In addition, the top 3 predictive drug combinations have been approved by the public. The new model is expected that this model can be extended to predict the use of combination drugs for other diseases to find combinations of drugs with potential clinical benefits.

Knowledge-infused Learning for Entity Prediction in Driving Scenes.

Scene understanding is a key technical challenge within the autonomous driving domain. It requires a deep semantic understanding of the entities and relations found within complex physical and social environments that is both accurate and complete. In practice, this can be accomplished by representing entities in a scene and their relations as a knowledge graph (KG). This scene knowledge graph may then be utilized for the task of entity prediction, leading to improved scene understanding. In this paper, we will define and formalize this problem as Knowledge-based Entity Prediction (KEP). KEP aims to improve scene understanding by predicting potentially unrecognized entities by leveraging heterogeneous, high-level semantic knowledge of driving scenes. An innovative neuro-symbolic solution for KEP is presented, based on knowledge-infused learning, which 1) introduces a dataset agnostic ontology to describe driving scenes, 2) uses an expressive, holistic representation of scenes with knowledge graphs, and 3) proposes an effective, non-standard mapping of the KEP problem to the problem of link prediction (LP) using knowledge-graph embeddings (KGE). Using real, complex and high-quality data from urban driving scenes, we demonstrate its effectiveness by showing that the missing entities may be predicted with high precision (0.87 Hits@1) while significantly outperforming the non-semantic/rule-based baselines.

Features fusion based link prediction in dynamic neworks

Link prediction is the process of detecting new or missing links in any network. This work focuses on link prediction in dynamic networks. One of the major challenges of the link prediction problem is to achieve high accuracy. One of the well-known methods for link prediction is the similarity-based method, which uses similarity-based score. The three widely used similarity-based indices are Local (L), Global (G), and Quasi-local (Q) for calculating similarity scores. In this work, we have proposed a model, namely the LGQ model. In our LGQ model, the wide categories of indices are used in different combinations(L, G, Q, LG, LQ, GQ, LGQ) for feature set generation that can be used with various machine learning techniques for link prediction. In local similarity indices, we have considered Common Neighbors (CN), Adamic/Adar Index (AA), Jaccard Coefficient (JC), and Preferential Attachment (PA). In global similarity indices, we have considered cos+, Average Commute Time (ACT), Shortest Path (SP), MFI (Matrix Forest Index), and in quasi-local indices, local path Index (LP), Path of Length 3 (L3). We have tested and validated our proposed model by conducting numerical simulations on six well-known dynamic network datasets. The results demonstrate that the proposed LGQ model and its variations outperform several baseline algorithms in terms of AUPR, F1 score, Balanced accuracy score (BAC), and AUC. We also demonstrate and contrast the preference of Neural Network and Xgboost based prediction models for different variations of proposed feature sets.

Automatic detection of COVID-19 vaccine misinformation with graph link prediction

Enormous hope in the efficacy of vaccines became recently a successful reality in the fight against the COVID-19 pandemic. However, vaccine hesitancy, fueled by exposure to social media misinformation about COVID-19 vaccines became a major hurdle. Therefore, it is essential to automatically detect where misinformation about COVID-19 vaccines on social media is spread and what kind of misinformation is discussed, such that inoculation interventions can be delivered at the right time and in the right place, in addition to interventions designed to address vaccine hesitancy. This paper is addressing the first step in tackling hesitancy against COVID-19 vaccines, namely the automatic detection of known misinformation about the vaccines on Twitter, the social media platform that has the highest volume of conversations about COVID-19 and its vaccines. We present CoVaxLies, a new dataset of tweets judged relevant to several misinformation targets about COVID-19 vaccines on which a novel method of detecting misinformation was developed. Our method organizes CoVaxLies in a Misinformation Knowledge Graph as it casts misinformation detection as a graph link prediction problem. The misinformation detection method detailed in this paper takes advantage of the link scoring functions provided by several knowledge embedding methods. The experimental results demonstrate the superiority of this method when compared with classification-based methods, widely used currently.

Relation-aware Heterogeneous Graph Transformer based drug repurposing

Drug repurposing refers to discovery of new medical instructions for existing chemical drugs, which has great pharmaceutical significance. Recently, large-scale biological datasets are increasingly available, and many graph neural network (GNN) based methods for drug repurposing have been developed. These methods often deem drug repurposing as a link prediction problem, which mines features of biological data to identify drug–disease associations (i.e., drug–disease links). Due to heterogeneity of data, we need to deeply explore heterogeneous information of biological network for drug repurposing. In this paper, we propose a Relation-aware Heterogeneous Graph Transformer (RHGT) model to capture heterogeneous information for drug repurposing. We first construct a drug–gene–disease interactive network-based on biological data, and then propose a three-level network embedding model, which learns network embeddings at fine-grained subtype-level, node-level and coarse-grained edge-level, respectively. The output of subtype-level is the input of node-level and edge-level, and the output of node-level is the input of edge level. We get edge embeddings at edge-level, which integrates edge type embeddings and node embeddings. We deem that in this way, characteristics of drug–gene–disease interactive network can be captured more comprehensively. Finally, we identify drug–disease associations (i.e., drug–disease links) based on the relationship between drug–gene edge embeddings and gene–disease edge embeddings. Experimental results show that our model performs better than other state-of-the-art graph neural network methods, which validates effectiveness of the proposed model.

Mining relationships between performance of link prediction algorithms and network structure

The numerous link prediction algorithms proposed by the network science researchers demonstrate their creativity in this hot topic. However, various algorithms together with the miscellaneous real-world networks put much difficulty on the choice of algorithm when coping with a new network. In this paper, we try to provide some elementary rules through mining the relationships between network structure features and the algorithm mechanisms. We discovered some principles indicating clustering coefficients influences on the prediction accuracy of structure-based algorithms. Besides, our experiment results present some interesting phenomenon neglected previously. The results and discussions may help us understand the link prediction problem better and further.

Community-guided link prediction in multiplex networks

Multiplex link prediction is the problem of finding missing links between nodes based on information from other layers. Although the link prediction problem in the online social networks is studied comprehensively, most approaches only employ internal features of the under prediction layer and do not consider additional link information from other networks. Also, many existing link prediction techniques are only based on the extracted information from links or nodes. However, the information flow in many real-world systems like social networks is considered as collaborative relations on correlated groups as an alternative for individual relations. In this research, we have proposed a Community-guided Link Prediction based on External Similarity (CLPES) method for multiplex networks in which, beside nodes and links information, community information is also employed. In our proposed method, we used an evolutionary algorithm (MOEA/D-TS) for specifying the community structure of the desired network. Next, the incorporation of internal features of each layer with a new external similarity metric (ExSim) obtains the final values for the likelihood of link formation in the network. Experiments on various real-world multiplex networks prove the capability of the proposed CLPES method for producing improved results and its superiority against other link prediction algorithms.

Abuse detection in healthcare insurance with disease-treatment network embedding

Abuse in healthcare insurance refers to a medical service or practice inconsistent with the generally accepted sound fiscal practices, such as overtreatment or overcharging. These types of abuses may lead to prescriptions that do not meet the criteria for medical stability. On the other hand, abuse may incur unnecessary costs by deliberately executing gratuitous treatments. In efforts to detect and prevent abuse, insurance companies hire medical professionals to manually examine the legitimacy of claim filings. It is, however, very costly in terms of labor and time to review all of the claims given the exploding amount of filings. In this light, there are growing interests for employing data mining techniques to automatically detect abusive claims or providers showing an abnormal billing pattern. Unfortunately, most of these models do not consider the disease-treatment information explicitly. In order for detection models to properly address the issues rising from individual drugs with similar efficacy, it is absolutely essential to account for the relationship between diseases and treatments during the learning process. In this paper, we propose a network-based approach which assesses the relationship between the diseases and treatments when detecting abuse from claim filings. Our proposed model consists of three stages. During the first stage, a disease-treatment network is constructed based on information extracted from the claim filings. Since the association between diseases and treatments is not explicitly expressed on these filings, we infer the disease-treatment relationship by computing the relative risk (RR). Second stage involves selecting the best graph embedding method from several candidates. We select the best method by comparing performances on link prediction. At the final stage, we solve a link prediction problem as a vehicle to detecting overtreatments. If our link prediction model predicts links to be nonexistent for all of the diseases and treatments listed in a given claim, then the claim is classified as an overtreatment case. We test the proposed model using the real-world claim data and showed that the proposed method classifies the treatment well which does not explicitly exist in the training network.

LPbyCD: a new scalable and interpretable approach for Link Prediction via Community Detection in bipartite networks

Many aspects from real life with bi-relational structure can be modeled as bipartite networks. This modeling allows the use of some standard solutions for prediction and/or recommendation of new relations between objects in such networks. In this work, we combine an existing bipartite local models method with approaches for link prediction from communities to address the link prediction problem in this type of networks. The motivation of this work stems from the importance of an application task, drug–target interaction prediction. Searching valid drug candidates for a given biological target is an essential part of modern drug development. We model the problem as link prediction in a bipartite multi-layer network, which helps to aggregate different sources of information into one single structure and as a result improves the quality of link prediction. We adapt existing community measures for link prediction to the case of bipartite multi-layer networks, propose alternative ways for exploiting communities, and show experimentally that our approach is competitive with the state-of-the-art. We also demonstrate the scalability of our approach and assess interpretability. Additional evaluations on data of a different origin than drug–target interactions demonstrate the genericness of the proposed approach.

Hierarchical Attention Link Prediction Neural Network

In this paper, a novel end-to-end neural link prediction model, named Hierarchical Attention Link Prediction Neural Network (HalpNet), is proposed. HalpNet comprehensively explores neighborhood information, which has proved important for link prediction, via the core component, i.e., hierarchical attention mechanism. The proposed hierarchical attention mechanism consists of two neural attention layers, modeling crucial structure information at node level and subgraph level, respectively. At node level, a structure-preserving attention is developed to preserve structure features of each node in the neighborhood subgraph. Based on the latent node features, at subgraph level, a structure-aggregating attention is designed to learn how important that each node in the subgraph is for the linkage of the target node pair and aggregate node features with learned attentions as a comprehensive subgraph representation. Given this expressive representation of neighborhood subgraph, HalpNet is able to predict link score of target node pair effectively. We evaluate HalpNet on 8 benchmark datasets against 14 popular and state-of-the-art approaches. The experimental results demonstrate its significant superiority and wide applicability on link prediction problem.

Multiplex network embedding for implicit sentiment analysis

As one of the hot research directions in natural language processing, sentiment analysis has received continuous and extensive attention. Different from explicit sentiment words indicating sentiment polarity, implicit sentiment analysis is a more challenging problem due to the lack of sentiment words, which makes it inadequate to use traditional sentiment analysis method to judge the polarity of implicit sentiment. This paper takes sentiment analysis as a special sign link prediction problem, which is different from traditional text-based method. In particular, by performing the word graph-based text level information embedding and heterogeneous social network information embedding (i.e. user social relationship network embedding, and user-entity sentiment network embedding), the proposed scheme learns the highly nonlinear representations of network nodes, explores early fusion method to combine the strength of these two types of embedding modeling, optimizes all parameters simultaneously and creates enhanced context representations, leading to better capture of implicit sentiment polarity. The proposed method has been examined on real-world dataset, for implicit sentiment link prediction task. The experimental results demonstrate that the proposed method outperforms state-of-the-art schemes, including LINE, node2vec, and SDNE, by 20.2%, 19.8%, and 14.0%, respectively, on accuracy, and achieves at least 14% gains on AUROC. For sentiment analysis accuracy, the proposed method achieves AUROC of 80.6% and accuracy of 78.3%, which is at least 31% better than other models. This work can provide useful guidance on the implicit sentiment analysis.

A A Survey of the Link Prediction on Static and Temporal Knowledge Graph

Link prediction in knowledge graphs gradually plays an essential role in the field of research and application. Through detecting latent connections, we can refine the knowledge in the graph, discover interesting relationships, answer user questions or make item suggestions. In this paper, we conduct a survey of the methods that are currently achieving good results in link prediction. Specially, we perform surveys on both static and temporal graphs. First, we divide the algorithms into groups based on the characteristic representation of entities and relations. After that, we describe the original idea and analyze the key improvements. In each group, comparisons and investigation on the pros and cons of each method as well as their applications are made. Based on that, the correlation of the two graph types in link prediction is drawn. Finally, from the overview of the link prediction problem, we propose some directions to improve the models for future studies.

Transfer learning based attack detection for wireless communication networks

AbstractWireless communication network attack detection is important in the field of wireless communication network data mining. Transfer learning is a powerful tool to solve the problem of link prediction in unlabeled wireless networks. In the feature mapping process of transfer learning, the loss of source domain information is very serious, which leads to the loss of knowledge in inverse mapping. To obtain more specific and complete information in the source domain, a link prediction method based on the distribution function fitting method of domain orthogonal term family is proposed. The key samples in the source domain are used to construct a family of orthogonal polynomials, and the least square approximation of the distribution function is obtained by using the family of orthogonal polynomials. The feature vector set of the key samples is extracted from the source domain, and the pseudo label is assigned to the target training set, so that the link classifier can complete the supervised link prediction task. In addition, an advanced performance of TLAD is obtained by comparing with other state‐of‐the‐art models in practice.

Identifying potential association on gene-disease network via dual hypergraph regularized least squares

BackgroundIdentifying potential associations between genes and diseases via biomedical experiments must be the time-consuming and expensive research works. The computational technologies based on machine learning models have been widely utilized to explore genetic information related to complex diseases. Importantly, the gene-disease association detection can be defined as the link prediction problem in bipartite network. However, many existing methods do not utilize multiple sources of biological information; Additionally, they do not extract higher-order relationships among genes and diseases.ResultsIn this study, we propose a novel method called Dual Hypergraph Regularized Least Squares (DHRLS) with Centered Kernel Alignment-based Multiple Kernel Learning (CKA-MKL), in order to detect all potential gene-disease associations. First, we construct multiple kernels based on various biological data sources in gene and disease spaces respectively. After that, we use CAK-MKL to obtain the optimal kernels in the two spaces respectively. To specific, hypergraph can be employed to establish higher-order relationships. Finally, our DHRLS model is solved by the Alternating Least squares algorithm (ALSA), for predicting gene-disease associations.ConclusionComparing with many outstanding prediction tools, DHRLS achieves best performance on gene-disease associations network under two types of cross validation. To verify robustness, our proposed approach has excellent prediction performance on six real-world networks. Our research work can effectively discover potential disease-associated genes and provide guidance for the follow-up verification methods of complex diseases.