Link Prediction Problem Research Articles

Link prediction based on sampling in complex networks

The link prediction problem has received extensive attention in fields such as sociology, anthropology, information science, and computer science. In many practical applications, we only need to predict the potential links between the vertices of interest, instead of predicting all of the links in a complex network. In this paper, we propose a fast similarity based approach for predicting the links related to a given node. We construct a path set connected to the given node by a random walk. The similarity score is computed within a small sub-graph formed by the path set connected to the given node, which significantly reduces the computation time. By choosing the appropriate number of sampled paths, we can restrict the error of the estimated similarities within a given threshold. Our experimental results on a number of real networks indicate that the algorithm proposed in this paper can obtain accurate results in less time than existing methods.

Drug Response Prediction as a Link Prediction Problem

Drug response prediction is a well-studied problem in which the molecular profile of a given sample is used to predict the effect of a given drug on that sample. Effective solutions to this problem hold the key for precision medicine. In cancer research, genomic data from cell lines are often utilized as features to develop machine learning models predictive of drug response. Molecular networks provide a functional context for the integration of genomic features, thereby resulting in robust and reproducible predictive models. However, inclusion of network data increases dimensionality and poses additional challenges for common machine learning tasks. To overcome these challenges, we here formulate drug response prediction as a link prediction problem. For this purpose, we represent drug response data for a large cohort of cell lines as a heterogeneous network. Using this network, we compute “network profiles” for cell lines and drugs. We then use the associations between these profiles to predict links between drugs and cell lines. Through leave-one-out cross validation and cross-classification on independent datasets, we show that this approach leads to accurate and reproducible classification of sensitive and resistant cell line-drug pairs, with 85% accuracy. We also examine the biological relevance of the network profiles.

Metric Learning Combining With Boosting for User Distance Measure in Multiple Social Networks

How to model user distance from multiple social networks is an important challenge. People often simultaneously appear in multiple social networks that can provide complementary services. Thus, knowledge from different social networks can help overcome the data sparseness problem. However, the knowledge cannot be directly obtained due to that they are from different social networks. To solve this problem, we construct an adaptive model to learn user distance in multiple social networks via combining distance metric learning and boosting technologies. The basic idea of our model is to embed related social networks into a potential feature space, while retaining the topologies of social networks. To get the solution to our model, we formulate it as a convex optimization problem. Moreover, we propose an adaptive user distance measure algorithm whose time complexity is linear with the number of the links. We verify the feasibility and effectiveness of our model on the link prediction problem. Experiments on two real large-scale data sets demonstrate that our method outperforms the compared methods. To the best of our knowledge, the joint learning of metric learning with boosting is first studied in multiple social networks.

A utility-based link prediction method in social networks

Link prediction is a fundamental task in social networks, with the goal of estimating the likelihood of a link between each node pair. It can be applied in many situations, such as friend discovery on social media platforms or co-author recommendations in collaboration networks. Compared to the numerous traditional methods, this paper introduces utility analysis to the link prediction method by considering that individual preferences are the main reason behind the decision to form links, and meanwhile it also focuses on the meeting process that is a latent variable during the process of forming links. Accordingly, the link prediction problem is formulated as a machine learning process with latent variables; therefore, an Expectation–Maximization (EM, for short) algorithm is adopted and further developed to cope with the estimation problem. The performance of the present method is tested both on synthetic networks and on real-world datasets from social media networks and collaboration networks. All of the computational results illustrate that the proposed method yields more satisfying link prediction results than the selected benchmarks, and in particular, logistic regression, as a special case of the proposed method, provides the lower boundary of the likelihood function.

A Survey of Link Prediction in Complex Networks

Networks have become increasingly important to model complex systems composed of interacting elements. Network data mining has a large number of applications in many disciplines including protein-protein interaction networks, social networks, transportation networks, and telecommunication networks. Different empirical studies have shown that it is possible to predict new relationships between elements attending to the topology of the network and the properties of its elements. The problem of predicting new relationships in networks is called link prediction. Link prediction aims to infer the behavior of the network link formation process by predicting missed or future relationships based on currently observed connections. It has become an attractive area of study since it allows us to predict how networks will evolve. In this survey, we will review the general-purpose techniques at the heart of the link prediction problem, which can be complemented by domain-specific heuristic methods in practice.

Predicting POI Visits in a Heterogeneous Location-Based Social Network

A point of interest (POI) is a specific location that people may find useful or interesting, such as restaurants, stores, attractions, and hotels. With the recent proliferation of location-based social networks (LBSN), numerous users gather to interact and share information on various POIs. POI recommendations have become a crucial issue because it not only helps users to learn about new places but also gives LBSN providers chances to post POI advertisements. As we utilize a heterogeneous information network to represent an LBSN in this work, POI recommendations are remodeled as a link prediction problem, which is significant in the field of social network analysis. Moreover, we propose to utilize the meta-path-based approach to extract implicit but potentially useful relationships between a user and a POI. Then, the extracted topological features are used to construct a prediction model with appropriate data classification techniques. In our experiments, the Yelp dataset is utilized as our testbed for performance evaluation purposes. The results show that our prediction model is of good prediction quality in practical applications.

An Approach to Cold-Start Link Prediction: Establishing Connections between Non-Topological and Topological Information

Cold-start link prediction is a term for information starved link prediction where little or no topological information is present to guide the determination of whether links to a node will form. Due to the lack of topological information, traditional topology-based link prediction methods cannot be applied to solve the cold-start link prediction problem. Therefore, an effective approach is presented through establishing connections between non-topological and topological information. In the approach, topological information is first extracted by a latent-feature representation model, then a logistic model is proposed to establish the connections between topological and non-topological information, and finally the linking possibility between cold-start users and existing users is calculated. Experiments with three types of real-world social networks Weibo, Facebook, and Twitter show that the proposed approach is more effective in solving the cold-start link prediction problem and establishing connections between topological and non-topological information.

A multilayered approach for link prediction in heterogeneous complex networks

Link prediction problem is a difficult task in complex networks due to (i) network size and sparsity, and (ii) extracting efficient similarity measures between node pairs. Although many link predictors have been proposed in the literatures, most of them are presented in homogeneous networks. The heterogeneity of node types adds a new challenge to link prediction problem. A meta-structure known as meta-path has been recently proposed to overcome the heterogeneity challenges. However, defining and generating good meta-paths as well as obtaining high prediction accuracy are still open issues. In this paper, a multilayered approach has been proposed in which a heterogeneous system is modeled as a multilayered complex network. Then, by exploring the network layers with different semantics, a set of meta-paths is generated. Extracting a number of topological features for each meta-path, a number of link predictors is learned which are aggregated to build the final link predictor. The experimental results on a bibliography network (DBLP) show that the proposed approach obtains higher accuracy in comparison with popular heterogeneous proximity measures.

REDUCING ATTRIBUTES of FACEBOOK USERS USING ROUGH SET THEORY

Using social networks have become one of the daily activities that billions of peoples around the world do. So, great research efforts had been done to analyze and understand these virtual communities. Among other things, link prediction is a paramount task to analyze and understand these social networks. In this paper, we investigate link prediction problem using rough set theory to discard the irrelevant attributes that could be found in the profiles of Facebook users and the proposed workinduces accuracy 97.79%.

Sampling-based algorithm for link prediction in temporal networks

The problem of link prediction in temporal networks has attracted considerable recent attention from various domains, such as sociology, anthropology, information science, and computer science. In this paper, we propose a fast similarity-based method to predict the potential links in temporal networks. In this method, we first combine the snapshots of the temporal network into a weighted graph. A proper damping factor is used to assign greater importance to more recent snapshots. Then, we construct a sub-graph centered at each node in the weighted graph by a random walk from the node. The sub-graph constructed consists of a set of paths starting from the given node. Because the similarity score is computed within such small sub-graphs centered at each node, the algorithm can greatly reduce the computation time. By choosing a proper number of sampled paths, we can restrict the error of the estimated similarities within a given threshold. While other random walk-based algorithms require O(n3) time for a network with n nodes, the computation time of our algorithm is O(n2), which is the lowest time complexity of a similarity-based link prediction algorithm. Moreover, because the proposed method integrates temporal and global topological information in the network, it can yield more accurate results. The experimental results on real networks show that our algorithm demonstrates the best or comparable quality results in less time than other methods.

A multilayer approach to multiplexity and link prediction in online geo-social networks.

Online social systems are multiplex in nature as multiple links may exist between the same two users across different social media. In this work, we study the geo-social properties of multiplex links, spanning more than one social network and apply their structural and interaction features to the problem of link prediction across social networking services. Exploring the intersection of two popular online platforms - Twitter and location-based social network Foursquare - we represent the two together as a composite multilayer online social network, where each platform represents a layer in the network. We find that pairs of users connected on both services, have greater neighbourhood similarity and are more similar in terms of their social and spatial properties on both platforms in comparison with pairs who are connected on just one of the social networks. Our evaluation, which aims to shed light on the implications of multiplexity for the link generation process, shows that we can successfully predict links across social networking services. In addition, we also show how combining information from multiple heterogeneous networks in a multilayer configuration can provide new insights into user interactions on online social networks, and can significantly improve link prediction systems with valuable applications to social bootstrapping and friend recommendations.

Improving probabilistic inference in graphical models with determinism and cycles

Many important real-world applications of machine learning, statistical physics, constraint programming and information theory can be formulated using graphical models that involve determinism and cycles. Accurate and efficient inference and training of such graphical models remains a key challenge. Markov logic networks (MLNs) have recently emerged as a popular framework for expressing a number of problems which exhibit these properties. While loopy belief propagation (LBP) can be an effective solution in some cases; unfortunately, when both determinism and cycles are present, LBP frequently fails to converge or converges to inaccurate results. As such, sampling based algorithms have been found to be more effective and are more popular for general inference tasks in MLNs. In this paper, we introduce Generalized arc-consistency Expectation Maximization Message-Passing (GEM-MP), a novel message-passing approach to inference in an extended factor graph that combines constraint programming techniques with variational methods. We focus our experiments on Markov logic and Ising models but the method is applicable to graphical models in general. In contrast to LBP, GEM-MP formulates the message-passing structure as steps of variational expectation maximization. Moreover, in the algorithm we leverage the local structures in the factor graph by using generalized arc consistency when performing a variational mean-field approximation. Thus each such update increases a lower bound on the model evidence. Our experiments on Ising grids, entity resolution and link prediction problems demonstrate the accuracy and convergence of GEM-MP over existing state-of-the-art inference algorithms such as MC-SAT, LBP, and Gibbs sampling, as well as convergent message passing algorithms such as the concave---convex procedure, residual BP, and the L2-convex method.

Truncated Kernel Projection Machine for Link Prediction

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.

Followee recommendation in Twitter using fuzzy link prediction

AbstractIn social networking sites, it is useful to receive recommendations about whom to contact or follow. These recommendations not only allow to establish connections with people one might already know in real life but also with people or users that have similar interests or are potentially interesting. We propose an approach that tackles contact (followee) recommendation in Twitter by means of fuzzy logic. This fuzzy approach handles recommendation as a link prediction problem and uses three types of similarity between a pair of users: tweet similarity, followee id similarity, and followee tweet similarity. These similarities are calculated by extracting user profiles. These profiles are, in turn, obtained by considering Twitter as a heterogeneous information network. To test our approach, we crawled a repository of 6000 users and two million tweets, and we measured accuracy by comparing our results with the actual followee lists of the users. These results, which are also compared against the results given by state‐of‐the‐art methods, show a high accuracy. Other advantages of the fuzzy system include a self‐explanatory capability and the ability to produce a non‐binary friendship value.

Efficient energy-based embedding models for link prediction in knowledge graphs

We focus on the problem of link prediction in Knowledge Graphs, with the goal of discovering new facts. To this purpose, Energy-Based Models for Knowledge Graphs that embed entities and relations in continuous vector spaces have been largely used. The main limitation in their applicability lies in the parameter learning phase, which may require a large amount of time for converging to optimal solutions. In this article, we first propose an unified view on different Energy-Based Embedding Models. Hence, for improving the model training phase, we propose the adoption of adaptive learning rates. We show that, by adopting adaptive learning rates during training, we can improve the efficiency of the parameter learning process by an order of magnitude, while leading to more accurate link prediction models in a significantly lower number of iterations. We extensively evaluate the proposed learning procedure on a variety of new models: our result show a significant improvement over state-of-the-art link prediction methods on two large Knowledge Graphs, namely WordNet and Freebase.

Link Prediction in Criminal Networks: A Tool for Criminal Intelligence Analysis.

The problem of link prediction has recently received increasing attention from scholars in network science. In social network analysis, one of its aims is to recover missing links, namely connections among actors which are likely to exist but have not been reported because data are incomplete or subject to various types of uncertainty. In the field of criminal investigations, problems of incomplete information are encountered almost by definition, given the obvious anti-detection strategies set up by criminals and the limited investigative resources. In this paper, we work on a specific dataset obtained from a real investigation, and we propose a strategy to identify missing links in a criminal network on the basis of the topological analysis of the links classified as marginal, i.e. removed during the investigation procedure. The main assumption is that missing links should have opposite features with respect to marginal ones. Measures of node similarity turn out to provide the best characterization in this sense. The inspection of the judicial source documents confirms that the predicted links, in most instances, do relate actors with large likelihood of co-participation in illicit activities.

Link prediction based on temporal similarity metrics using continuous action set learning automata

Abstract Link prediction is a social network research area that tries to predict future links using network structure. The main approaches in this area are based on predicting future links using network structure at a specific period, without considering the links behavior through different periods. For example, a common traditional approach in link prediction calculates a chosen similarity metric for each non-connected link and outputs the links with higher similarity scores as the prediction result. In this paper, we propose a new link prediction method based on temporal similarity metrics and Continuous Action set Learning Automata (CALA). The proposed method takes advantage of using different similarity metrics as well as different time periods. In the proposed algorithm, we try to model the link prediction problem as a noisy optimization problem and use a team of CALAs to solve the noisy optimization problem. CALA is a reinforcement based optimization tool which tries to learn the optimal behavior from the environment feedbacks. To determine the importance of different periods and similarity metrics on the prediction result, we define a coefficient for each of different periods and similarity metrics and use a CALA for each coefficient. Each CALA tries to learn the true value of the corresponding coefficient. Final link prediction is obtained from a combination of different similarity metrics in different times based on the obtained coefficients. The link prediction results reported here show satisfactory of the proposed method for some social network data sets.

A class-based link prediction using Distance Dependent Chinese Restaurant Process

One of the important tasks in relational data analysis is link prediction which has been successfully applied on many applications such as bioinformatics, information retrieval, etc. The link prediction is defined as predicting the existence or absence of edges between nodes of a network. In this paper, we propose a novel method for link prediction based on Distance Dependent Chinese Restaurant Process (DDCRP) model which enables us to utilize the information of the topological structure of the network such as shortest path and connectivity of the nodes. We also propose a new Gibbs sampling algorithm for computing the posterior distribution of the hidden variables based on the training data. Experimental results on three real-world datasets show the superiority of the proposed method over other probabilistic models for link prediction problem.

Link prediction based on path entropy

Information theory has been taken as a prospective tool for quantifying the complexity of complex networks. In this paper, first we study the information entropy or uncertainty of a path using the information theory. After that, we apply the path entropy to the link prediction problem in real-world networks. Specifically, we propose a new similarity index, namely Path Entropy (PE) index, which considers the information entropies of shortest paths between node pairs with penalization to long paths. Empirical experiments demonstrate that PE index outperforms the mainstream of link predictors.

Combining Two and Three-Way Embedding Models for Link Prediction in Knowledge Bases

This paper tackles the problem of endogenous link prediction for knowledge base completion. Knowledge bases can be represented as directed graphs whose nodes correspond to entities and edges to relationships. Previous attempts either consist of powerful systems with high capacity to model complex connectivity patterns, which unfortunately usually end up overfitting on rare relationships, or in approaches that trade capacity for simplicity in order to fairly model all relationships, frequent or not. In this paper, we propose Tatec, a happy medium obtained by complementing a high-capacity model with a simpler one, both pre-trained separately and then combined. We present several variants of this model with different kinds of regularization and combination strategies and show that this approach outperforms existing methods on different types of relationships by achieving state-of-the-art results on four benchmarks of the literature.