Link prediction in complex networks based on cluster information

Encyclopedia of Business Analytics and Optimization, 2014

One of the major fields of study in the analysis and mining of social networks is the study of social network patterns and development. Link prediction has attracted extensive attention from the scientist in the network science. Despite incredible endeavors to create modern prediction algorithms that can give way better prediction precision in social network topologies that are extricated from complex datasets approximately clients, their exercises, and connections. But most of the recent approaches have not been broadly evaluated. This paper focuses on the missing link prediction between nodes in a systematic way by formalizing and developing approaches about future interactions that can be extracted from network topology for analyzing the proximity measures of nodes, links, and their attributes in a network. The main assumption of this paper is that the missing links must have opposite features concerning the original links. It turns out that feature engineering via measured the node or edge attributes and network structures based provided the best characterization in this sense. Some of network topology-based prediction methods are Common Neighbors, Sørensen Index, Hub Depressed Index, Adamic Index, Salton's cosine formula, spectral clustering, etc. Experimental results using AUC and ROC proved that based on Network topology and previous information may be more or less difficult to accurately predicted future connections so, there is no single clear winner among similarity metrics.

Scientific Reports, 2021

The entities of real-world networks are connected via different types of connections (i.e., layers). The task of link prediction in multiplex networks is about finding missing connections based on both intra-layer and inter-layer correlations. Our observations confirm that in a wide range of real-world multiplex networks, from social to biological and technological, a positive correlation exists between connection probability in one layer and similarity in other layers. Accordingly, a similarity-based automatic general-purpose multiplex link prediction method-SimBins-is devised that quantifies the amount of connection uncertainty based on observed inter-layer correlations in a multiplex network. Moreover, SimBins enhances the prediction quality in the target layer by incorporating the effect of link overlap across layers. Applying SimBins to various datasets from diverse domains, our findings indicate that SimBins outperforms the compared methods (both baseline and state-of-the-art methods) in most instances when predicting links. Furthermore, it is discussed that SimBins imposes minor computational overhead to the base similarity measures making it a potentially fast method, suitable for large-scale multiplex networks. Link prediction has been an area of interest in the research of complex networks for over two decades 1 , studying the relationships between entities (nodes) in data represented as graphs. The main goal is to reveal the underlying truth behind emerging or missing connections between node pairs of a network. Link prediction methods have a wide range of applications, from discovery of latent and spurious interactions in biological networks (which is basically quite costly if performed in traditional methods) 2,3 to recommender systems 4,5 and better routing in wireless mobile networks 6. Numerous perspectives have been adopted to attack the problem of link prediction. According to similarity-based methods, similarity between nodes determines their likelihood of linkage. This approach is a result of assuming that two nodes are similar if they share many common features 7. A whole lot of nodes' features stay hidden (or are kept hidden intentionally) in real networks. Further, an interesting question is, despite the fact that a considerable amount of information is hidden in a network, what fraction of the truth can still be extracted by merely including structural features? That is one of the main drives to utilize structural similarity indices for link prediction. Several different classifications of similarity measures have been proposed, among all, classifying based on locality of indices is of great importance. To name a few, Common Neighbors (CN) 1 , Preferential Attachment (PA) 8 , Adamic-Adar (AA) 9 and Resource Allocation (RA) 10 are popular indices focusing mostly on nodes' structural features, each with unique characteristics. Even though these indexes are simple, they are popular because of their low computational cost and reasonable prediction performance. On the other hand, global indices take features of the whole network structure into account, tolerating higher cost of computation, usually in favor of more accurate information. Take length of paths between pairs of nodes for instance, which the well-known Katz 11 index operates on. Average Commute Time (ACT) 1 and PageRank 12 are some other notable global indices. In between lie the quasi-local methods which are able to combine properties from both local and global indices, meaning they include global information, but their computational complexity is similar to that of local methods, such as the Local Path (LP) 13 index and Local Random Walk (LRW) 14. For more detailed information on these similarity indices (also described as unsupervised methods in the literature 15), readers are advised to refer to 16. Some researchers have tackled the link prediction problem using the ideas of information theory. These works are based on the fact that similarity of node pairs can be written in term of the uncertainty of their connectivity. At the beginning, the uncertainty of connectivity can be estimated based on priors. Later, all structures around the unconnected node pairs can be considered as evidences to reduce the level of uncertainty in connectedness

Journal of Parallel and Distributed Computing, 2017

Link prediction has become an important task, especially with the rise of large-scale, complex and dynamic networks. The emerging research area of network dynamics and evolution is directly related to predicting new interactions between objects, a possibility in the near future. Recent studies show that the precision of link prediction can be improved to a great extent by including community information in the prediction methods. As traditional community-based link prediction algorithms can run only on stand-alone computers, they are not well suited for most of the large networks. Graph parallelization can be one solution to such problems. Bulk Synchronous Parallel (BSP) programming model is a recently emerged framework for parallelizing graph algorithms. In this paper, we propose a hybrid similarity measure for link prediction in real world networks. We also propose a scalable method for community structure-based link prediction on large networks. This method uses a parallel label propagation algorithm for community detection and a parallel community information-based Adamic-Adar measure for link prediction. We have developed these algorithms using Bulk Synchronous Parallel programming model and tested them with large networks of various domains.

IETI Transactions on Data Analysis and Forecasting (iTDAF)

Link Prediction in Human Complex Networks aims to predict the missing, deleted, or future link formations. These complex networks are represented graphically, consisting of nodes and links, also referred to as vertices and edges, respectively. We employ Link Prediction techniques on four different human-related networks to determine the most effective methods in the Human Complex domain. The techniques utilized are similarity-based, primarily focused on determining the similarity score of each network. We select four algorithms that demonstrated superior results in other complex networks and implement them on human-related networks. Our goal is to predict links that have been removed from the network in order to evaluate the prediction accuracy of the applied techniques. To accomplish this, we convert the datasets into adjacency matrices and divide them into training and probe sets. The training session is then conducted, followed by the testing of the data. The selected techniques ...

Physica A: Statistical Mechanics and its Applications, 2016

Evaluation of link prediction methods is a hard task in very large complex networks because of the inhibitive computational cost. By setting a lower bound of the number of common neighbors (CN), we propose a new framework to efficiently and precisely evaluate the performances of CN-based similarity indices in link prediction for very large heterogeneous networks. Specifically, we propose a fast algorithm based on the parallel computing scheme to obtain all the node pairs with CN values larger than the lower bound. Furthermore, we propose a new measurement, called selfpredictability, to quantify the performance of the CN-based similarity indices in link prediction, which on the other side can indicate the link predictability of a network.

Proceedings of the 2014 CoNEXT on Student Workshop, 2014

The problem of link prediction has gained a lot of attention recently from the research community. It can be formalized as, given a snapshot of a social network at time t, can it be predicted which new connections among its members are likely to occur in the future at time t. Apart from analysing social networks, it has also found application in other domains e.g., information retrieval, bio-informatics and e-commerce. Topological information of the network, i.e. the information about the present nodes and links, can be used to predict the future links in the network. As an example, "Common neighbors" method is a trivial but efficient strategy for predicting the possibility of a link between a pair of nodes. Many variants of the common neighbors method have been proposed to address this problem. In this paper, we propose a novel strategy for predicting the missing links which also takes into account the number of links between two sets of uncommon neighbors of given nodes, in addition to their common neighbors.

In recent years, researchers from academic and industrial fields have become increasingly interested in social network data to extract meaningful information. This information is used in applications such as link prediction between people groups, community detection, protein module identification, etc. Therefore, the clustering technique has emerged as a solution to finding similarities between social network members. Recently, in most graph clustering solutions, the structural similarity of nodes is combined with their attribute similarity. The results of these solutions indicate that the graph's topological structure is more important. Since most social networks are sparse, these solutions often suffer from insufficient use of node features. This paper proposes a hybrid clustering approach for link prediction in heterogeneous information networks (HINs). In our approach, an adjacency vector is determined for each node until, in this vector, the weight of the direct edge or the...

Information Sciences, 2019

In recent years, most link prediction algorithms have focused on node similarity owing to the associated low computational complexity and promising accuracy. In addition to the classical CN-based indexes, some methods are based on network features, such as community structure, information dissemination, and intermediary influence probability, which are used for link prediction. Although these methods provide new insight into the problem and achieve improvements in certain respects, they also have some limitations. For example, it is difficult to predict target links if the number of interconnections between communities is small. However, most studies aim at achieving higher link prediction accuracy even though a network obtained by these methods is not optimized for information spread. Therefore, we propose a community-based link prediction method using an information diffusion algorithm (CLP-ID) to predict the missing links. First, we present a community detection algorithm that divides the network into clusters. Then, a novel algorithm based on information diffusion and community structure is proposed to predict target links. Finally, we conduct experiments on real-world networks to validate the performance of the proposed algorithm and compare it with state-of-the-art algorithms. Statistical tests demonstrate that the proposed method significantly differs from state-of-the-art algorithms.

Chinese Physics B, 2021

Network information mining is the study of the network topology, which may answer a large number of application-based questions towards the structural evolution and the function of a real system. The question can be related to how the real system evolves or how individuals interact with each other in social networks. Although the evolution of the real system may seem to be found regularly, capturing patterns on the whole process of evolution is not trivial. Link prediction is one of the most important technologies in network information mining, which can help us understand the evolution mechanism of real-life network. Link prediction aims to uncover missing links or quantify the likelihood of the emergence of nonexistent links from known network structures. Currently, widely existing methods of link prediction almost focus on short-path networks that usually have a myriad of close triangular structures. However, these algorithms on highly sparse or long-path networks have poor perfo...