Path-based extensions of local link prediction methods for complex networks

Aziz, Furqan; Gul, Haji; Uddin, Imad; Gkoutos, Georgios

doi:10.1038/s41598-020-76860-2

Cited by 27 publications

(19 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ey are the link prediction algorithm based on high-order path similarity by punishing the long path (HPS-LP) between the predicted node pairs [42], the link prediction method based on local and global structure information by measuring the relative entropy (RE) under the joint action of first-order and second-order neighbour information [43], the algorithm called HD that was proposed based on a new definition of global and quasilocal extensions of some commonly used local similarity indices [44], and the link prediction algorithm called MSLPA based on community preference information 10 Complexity by considering the network structure attributes and interest preferences of users as the dominant factors in a Twitter dataset [45]. For the two unweighted networks, comparison results of these algorithms based on the AUC evaluation index are shown in Figure 15.…”

Section: Algorithm Robustnessmentioning

confidence: 99%

Link Prediction Model for Weighted Networks Based on Evidence Theory and the Influence of Common Neighbours

2022

View full text Add to dashboard Cite

A link prediction model for weighted networks based on Dempster–Shafer (DS) evidence theory and the influence of common neighbours is proposed in this paper. First, three types of future common neighbours (FCNs) and their topological structures are proposed. Second, the concepts of endpoint weight influence, link weight influence, and high-strength node influence are introduced. Then, the similarity based on the impacts of current common neighbours (CCNs) and FCNs is defined, respectively. Finally, the two similarity indices are fused by the DS evidence theory. This model effectively integrates multisource information and completely exploits the influence of all CCNs and FCNs on similarity. Experiments are performed on 9 real and 40 simulation-weighted datasets, and these findings are compared with several classic algorithms. Results show that the proposed method has higher precision than other methods, which can achieve good performance in link prediction in weighted networks.

show abstract

Section: Algorithm Robustnessmentioning

confidence: 99%

Link Prediction Model for Weighted Networks Based on Evidence Theory and the Influence of Common Neighbours

2022

View full text Add to dashboard Cite

show abstract

“…In our recent work 25 , we have developed methods for defining path-based extensions of some of the local similarity indices for unipartite graphs. The advantage of our approach is that it not only considers paths with wider horizons, but also takes into account the contribution of each node on those paths.…”

Section: Path-based Similarity Indexmentioning

confidence: 99%

“…Lü et al 24 have proposed local path index that exploit of paths with wider horizon than common neighbour. Recently, we have proposed novel global and quasi-local indices of local link prediction methods and demonstrated their applications in real-world biological and social networks 25 .…”

mentioning

confidence: 99%

“…Our novel contributions include the development of pathbased similarity indices that can be used estimate the likelihood of existence of a missing link in a bipartite network. The approach adopted is based on our recent work 25 , where we have developed novel methods for predicting missing links in unipartite networks. Our method is efficient as it is based on paths of length 3, and we have validated its performance on four benchmark biomedical datasets.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Multimorbidity prediction using link prediction

Aziz

Cardoso

Bravo-Merodio

et al. 2021

Sci Rep

Self Cite

View full text Add to dashboard Cite

Multimorbidity, frequently associated with aging, can be operationally defined as the presence of two or more chronic conditions. Predicting the likelihood of a patient with multimorbidity to develop a further particular disease in the future is one of the key challenges in multimorbidity research. In this paper we are using a network-based approach to analyze multimorbidity data and develop methods for predicting diseases that a patient is likely to develop. The multimorbidity data is represented using a temporal bipartite network whose nodes represent patients and diseases and a link between these nodes indicates that the patient has been diagnosed with the disease. Disease prediction then is reduced to a problem of predicting those missing links in the network that are likely to appear in the future. We develop a novel link prediction method for static bipartite network and validate the performance of the method on benchmark datasets. By using a probabilistic framework, we then report on the development of a method for predicting future links in the network, where links are labelled with a time-stamp. We apply the proposed method to three different multimorbidity datasets and report its performance measured by different performance metrics including AUC, Precision, Recall, and F-Score.

show abstract

“…Common neighbor, Jaccard coefficient, preferential attachment, Adamic Adar index, resource allocation index algorithms are popular link prediction algorithms. In contrast, the path‐based algorithms 8 consider the path information between the nodes to find the similarity score. Some examples of the path based algorithms are Katz, Shortest path distance, random walk restart, and page rank.…”

Section: Introductionmentioning

confidence: 99%

Gravitational search algorithm‐driven missing links prediction in social networks

Singh

2022

Concurrency and Computation

View full text Add to dashboard Cite

The increased usage of online social networks in recent days attracts the attention of researchers. That makes analysis of social networks a significant concern. Link prediction is one of the problem of social networks, where links are predicted between users by analyzing the relationships between the nodes. As the size and usability of the social network grows, the accurate prediction of connections with limited public information is a challenge. This article introduces a novel approach for predicting missing links using a nature-inspired method, the gravitational search algorithm. We selected seven real-world networks and five widely used algorithms and also considered the sparsity of networks for the experimental evaluation of the proposed approach. The proposed GSA-driven algorithm has shown a better or comparable AUC result than the other standard algorithms on these networks.

show abstract

Path-based extensions of local link prediction methods for complex networks

Cited by 27 publications

References 45 publications

Link Prediction Model for Weighted Networks Based on Evidence Theory and the Influence of Common Neighbours

Link Prediction Model for Weighted Networks Based on Evidence Theory and the Influence of Common Neighbours

Multimorbidity prediction using link prediction

Gravitational search algorithm‐driven missing links prediction in social networks

Contact Info

Product

Resources

About