Conference Paper

Link prediction in weighted network based on reliable routes by machine learning approach

Authors:
  • University of Information Technology, Vietnam National University - Ho Chi Minh City
  • University of Information Technology, VNU-HCM, Vietnam
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

In data mining, link prediction for the networks is one of the areas of greatest interest today. Research achievements of link prediction problem can be applied in many fields such as study genetically transferred diseases, online marketing, ecommerce services, discover the structure of criminal networks, friend request in social networks... However, most of researchers focused on predicting the existence of links in previous studies of link prediction. The predicting weight of links has not been heavily researched. In this paper, we introduce an effective solution for weighted network. We propose a novel learning-based approach to weight prediction. Our approach presents the Topological Similarity Score (TSS) feature combined by six indices (CN, AA, RA, rWCN, rWAA, rWRA) to compute the similarity scores between nodes. We propose to utilize Support Vector Regression (SVR) with TSS feature to predict weights. All experiments were conducted on five data sets: Cel, USAir, Lesmis, ReHall, Netscience. Experimental results show that our approach can increase the weight prediction correlation coefficient by 70% over and reduce the error by 17%, compared to the baseline approach.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Snapshots, as a favorable model to exhibit dynamic behaviors, have been widely used in various application scenarios. By employing such representation, unsupervised learning methods are feasible to estimate the links at time t with the observed network structures at time [1, t − 1] [17][18][19]. Besides, statistical methods, such as Exponential Smoothing (EPS) [20] and Autoregressive Integrated Moving Average(ARIMA) [21], are also employed to predict temporal links with snapshots representation. However, Snapshots suffer from coarse-grained depiction of continuous changes, which probability result in poor predictive performance and misleading results [22]. ...
Article
Full-text available
Social network analysis is a multidisciplinary study covering informatics, mathematics, sociology, management, psychology, etc. Link prediction, as one of the fundamental studies with a variety of applications, has attracted increasing focus from scientific society. Traditional research based on graph theory has made numerous achievements, whereas suffering from incapability of dealing with dynamic behaviors and low predicting accuracy. Aiming at addressing the problem, this paper employs a diagonally symmetrical supra-adjacency matrix to represent the dynamic social networks, and proposes a temporal links prediction framework combining with an improved gravity model. Extensive experiments on several real-world datasets verified the superiority on competitors, which benefits recommending friends in social networks. It is of remarkable significance in revealing the evolutions in temporal networks and promoting considerable commercial interest for social applications.
... TCP is being improved by machine learning by training classifiers to discriminate between link errors and congestion using patterns in round-trip time statistics -which is very useful for wireless networks where flow control algorithms cannot distinguish losses caused by signal fading, a link-error, from losses caused by congestion, leading to systematic reduction of transfer rates when a packet is dropped [2]. Link prediction is evolving thanks to machine learning algorithms creating faster and more accurate weight predictions in graphs [3]. ...
Conference Paper
Full-text available
This paper shows that with bandwidth and roundtrip time statistics, data analytics can be used to classify three characteristic phenomena in wireless signal use: decreases in bandwidth due to signal over-saturation, signal attenuation due to increasing distance, and signal improvement due to decreasing distance. Using a K-Means algorithm, bandwidth and round-trip time trends were clustered correctly by signal loss type with a 99.98% accuracy rating with 10,000 validation samples.
Chapter
For the link prediction problem, it commonly estimates the similarity by different similarity metrics or machine learning prediction model. However, this paper proposes an algorithm, which is called Relation Pattern Deep Learning Classification (RPDLC) algorithm, based on two neighbor-based similarity metrics and convolution neural network. First, the RPDLC extracts the features for two nodes in a pair, which is calculated with neighbor-based metric and influence nodes. Second, the RPDLC combines the features of nodes to be a heat map for evaluating the similarity of the node’s relation pattern. Third, the RPDLC constructs the prediction model for predicting missing relationship by using convolution neural network architecture. In consequence, the contribution of this paper is purposed a novel approach for link prediction problem, which is used convolution neural network and features by relation pattern to construct a prediction model.
Article
Full-text available
Next-generation wireless networks are expected to support extremely high data rates and radically new applications, which require a new wireless radio technology paradigm. The challenge is that of assisting the radio in intelligent adaptive learning and decision making, so that the diverse requirements of next-generation wireless networks can be satisfied. Machine learning is one of the most promising artificial intelligence tools, conceived to support smart radio terminals. Future smart 5G mobile terminals are expected to autonomously access the most meritorious spectral bands with the aid of sophisticated spectral efficiency learning and inference, in order to control the transmission power, while relying on energy efficiency learning/inference and simultaneously adjusting the transmission protocols with the aid of quality of service learning/inference. Hence we briefly review the rudimentary concepts of machine learning and propose their employment in the compelling applications of 5G networks, including cognitive radios, massive MIMOs, femto/small cells, heterogeneous networks, smart grid, energy harvesting, device-todevice communications, and so on. Our goal is to assist the readers in refining the motivation, problem formulation, and methodology of powerful machine learning algorithms in the context of future networks in order to tap into hitherto unexplored applications and services.
Article
Full-text available
Link prediction aims to uncover missing links or predict the emergence of future relationships from the current network structure. Plenty of algorithms have been developed for link prediction in unweighted networks, but only a few have been extended to weighted networks. In this paper, we present what we call a "reliable-route method" to extend unweighted local similarity indices to weighted ones. Using these indices, we can predict both the existence of links and their weights. Experiments on various real-world networks suggest that our reliable-route weighted resource-allocation index performs noticeably better than others with respect to weight prediction. For existence prediction it is either the highest or very close to the highest. Further analysis shows a strong positive correlation between the clustering coefficient and prediction accuracy. Finally, we apply our method to the prediction of missing protein-protein interactions and their confidence scores from known PPI networks. Once again, our reliable-route method shows the highest accuracy.
Article
Full-text available
Various structural features of networks have been applied to develop link prediction methods. However, because different features highlight different aspects of network structural properties, it is very difficult to benefit from all of the features that might be available. In this paper, we investigate the role of network topology in predicting missing links from the perspective of information theory. In this way, the contributions of different structural features to link prediction are measured in terms of their values of information. Then, an information-theoretic model is proposed that is applicable to multiple structural features. Furthermore, we design a novel link prediction index, called Neighbor Set Information (NSI), based on the information-theoretic model. According to our experimental results, the NSI index performs well in real-world networks, compared with other typical proximity indices.
Article
Full-text available
Traditionally, numerous simple models such as preferential attachment have been put forward to reveal the evolution mechanisms of real networks. However, previous simulations show that real networks usually are driven by various features instead of single pure mechanism. To solve this problem, some pioneers proposed a few hybrid models of mixing multiple evolution mechanisms and tried to uncover the contributions of different mechanisms. In this paper, we introduce two methods which can tackle this problem: one is based on link prediction model, and the other is based on likelihood analysis. To examine the effectiveness, we generate plenty of artificial networks which can be controlled to follow multiple mechanisms with different weights, so that we can compare the estimated weights with the true values. The experimental results show the method based on likelihood analysis performs much better and gives very accurate estimations. At last, we apply this method to real networks to see how popularity and clustering co-evolve. We find most of them are affected by both popularity and clustering, but with quite different weights.
Article
Full-text available
Growth and remodelling impact the network topology of complex systems, yet a general theory explaining how new links arise between existing nodes has been lacking, and little is known about the topological properties that facilitate link-prediction. Here we investigate the extent to which the connectivity evolution of a network might be predicted by mere topological features. We show how a link/community-based strategy triggers substantial prediction improvements because it accounts for the singular topology of several real networks organised in multiple local communities - a tendency here named local-community-paradigm (LCP). We observe that LCP networks are mainly formed by weak interactions and characterise heterogeneous and dynamic systems that use self-organisation as a major adaptation strategy. These systems seem designed for global delivery of information and processing via multiple local modules. Conversely, non-LCP networks have steady architectures formed by strong interactions, and seem designed for systems in which information/energy storage is crucial.
Article
Full-text available
Support vector machines (SVMs) have improved generalization performance over other classical optimization techniques. Here, we introduce an SVM-based approach for linear array processing and beamforming. The development of a modified cost function is presented and it is shown how it can be applied to the problem of linear beamforming. Finally, comparison examples are included to show the validity of the new minimization approach.
Conference Paper
Full-text available
Link prediction is an important task in Social Network Analysis. This problem refers to predicting the emergence of future relationships between nodes in a social network. Our work focuses on a supervised machine learning strategy for link prediction. Here, the target attribute is a class label indicating the existence or absence of a link between a node pair. The predictor attributes are metrics computed from the network structure, describing the given pair. The majority of works for supervised prediction only considers unweighted networks. In this light, our aim is to investigate the relevance of using weights to improve supervised link prediction. Link weights express the `strength' of relationships and could bring useful information for prediction. However, the relevance of weights for unsupervised strategies of link prediction was not always verified (in some cases, the performance was even harmed). Our preliminary results on supervised prediction on a co-authorship network revealed satisfactory results when weights were considered, which encourage us for further analysis.
Conference Paper
Full-text available
A new regression technique based on Vapnik's concept of support vectors is introduced. We compare support vector regression (SVR) with a committee regression technique (bagging) based on regression trees and ridge regression done in feature space. On the basis of these experiments, it is expected that SVR will have advantages in high dimensionality space because SVR optimization does not depend on the dimensionality of the input space.
Conference Paper
Full-text available
As an integral part of reliable communication in wireless networks, effective link estimation is essential for routing protocols. However, due to the dynamic nature of wireless channels, accurate link quality estimation remains a challenging task. In this paper, we propose 4C, a novel link estimator that applies link quality prediction along with link estimation. Our approach is data-driven and consists of three steps: data collection, offline modeling and online prediction. The data collection step involves gathering link quality data, and based on our analysis of the data, we propose a set of guidelines for the amount of data to be collected in our experimental scenarios. The modeling step includes offline prediction model training and selection. We present three prediction models that utilize different machine learning methods, namely, naive Bayes classifier, logistic regression and artificial neural networks. Our models take a combination of PRR and the physical layer information, i.e., Received Signal Strength Indicator (RSSI), Signal to Noise Ratio (SNR) and Link Quality Indicator (LQI) as input, and output the success probability of delivering the next packet. From our analysis and experiments, we find that logistic regression works well among the three models with small computational cost. Finally, the third step involves the implementation of 4C, a receiver-initiated online link quality prediction module that computes the short temporal link quality. We conducted extensive experiments in the Motelab and our local indoor testbeds, as well as an outdoor deployment. Our results with single and multiple senders experiments show that with 4C, CTP improves the average cost of delivering a packet by 20% to 30%. In some cases, the improvement is larger than 45%.
Article
Full-text available
Common-neighbor-based method is simple yet effective to predict missing links, which assume that two nodes are more likely to be connected if they have more common neighbors. In such method, each common neighbor of two nodes contributes equally to the connection likelihood. In this Letter, we argue that different common neighbors may play different roles and thus lead to different contributions, and propose a local na\"{\i}ve Bayes model accordingly. Extensive experiments were carried out on eight real networks. Compared with the common-neighbor-based methods, the present method can provide more accurate predictions. Finally, we gave a detailed case study on the US air transportation network.
Article
Full-text available
Many models are put forward to mimic the evolution of real networked systems. A well-accepted way to judge the validity is to compare the modeling results with real networks subject to several structural features. Even for a specific real network, we cannot fairly evaluate the goodness of different models since there are too many structural features while there is no criterion to select and assign weights on them. Motivated by the studies on link prediction algorithms, we propose a unified method to evaluate the network models via the comparison of the likelihoods of the currently observed network driven by different models, with an assumption that the higher the likelihood is, the better the model is. We test our method on the real Internet at the Autonomous System (AS) level, and the results suggest that the Generalized Linear Preferential (GLP) model outperforms the Tel Aviv Network Generator (Tang), while both two models are better than the Barab\'asi-Albert (BA) and Erd\"os-R\'enyi (ER) models. Our method can be further applied in determining the optimal values of parameters that correspond to the maximal likelihood. Experiment indicates that the parameters obtained by our method can better capture the characters of newly-added nodes and links in the AS-level Internet than the original methods in the literature.
Article
Full-text available
The Support Vector Machine (SVM) is a widely used classifier in bioinformatics. Obtaining the best results with SVMs requires an understanding of their workings and the various ways a user can influence their accuracy. We provide the user with a basic understanding of the theory behind SVMs and focus on their use in practice. We describe the effect of the SVM parameters on the resulting classifier, how to select good values for those parameters, data normalization, factors that affect training time, and software for training SVMs.
Article
Full-text available
We investigated the relationship between friendships and suicidality among male and female adolescents. We analyzed friendship data on 13,465 adolescents from the National Longitudinal Survey of Adolescent Health to explore the relationship between friendship and suicidal ideation and suicide attempts. We controlled for known factors associated with suicidality. Having had a friend who committed suicide increased the likelihood of suicidal ideation and attempts for both boys and girls. Socially isolated females were more likely to have suicidal thoughts, as were females whose friends were not friends with each other. Among adolescents thinking about suicide, suicide attempts appear largely stochastic, with few consistent risk factors between boys and girls. The friendship environment affects suicidality for both boys and girls. Female adolescents' suicidal thoughts are significantly increased by social isolation and friendship patterns in which friends were not friends with each other.
Article
Full-text available
Many weighted scale-free networks are known to have a power-law correlation between strength and degree of nodes, which, however, has not been well explained. We investigate the dynamic behavior of resource-traffic flow on scale-free networks. The dynamical system will evolve into a kinetic equilibrium state, where the strength, defined by the amount of resource or traffic load, is correlated with the degree in a power-law form with tunable exponent. The analytical results agree well with simulations.
Article
Full-text available
A new regression technique based on concept of support vectors is introduced. We compare support vector regression with a committee regression technique (bagging) based on regression trees and ridge regression done in feature space. On the basis of these experiments, it is expected that SVR will have advantages in high dimensionality space because SVR optimization does not depend on the dimension&y of input space. This is a longer version of the paper appear in Advances in Neural Processing Systems 9 (proceedings of the 1996 conference)
Article
Thesupport-vector network is a new learning machine for two-group classification problems. The machine conceptually implements the following idea: input vectors are non-linearly mapped to a very high-dimension feature space. In this feature space a linear decision surface is constructed. Special properties of the decision surface ensures high generalization ability of the learning machine. The idea behind the support-vector network was previously implemented for the restricted case where the training data can be separated without errors. We here extend this result to non-separable training data.High generalization ability of support-vector networks utilizing polynomial input transformations is demonstrated. We also compare the performance of the support-vector network to various classical learning algorithms that all took part in a benchmark study of Optical Character Recognition.
Article
The control of communication networks is an important aspect from both the service provider and user points of view. There are several approaches to communication network control including game theory, genetic algorithms and Markov decision processes. Data mining methods have been successfully used to discover optimized solutions to this problem, and have the capability to learn the network behavior under different network conditions and during operation so that complete knowledge of the network behavior is not required a priori. This article identifies the concepts behind the idea of using data mining for communication network control, provides a structured survey of the results in this area, and discusses the guidelines for future applications.
Article
In social networks, link prediction predicts missing links in current networks and new or dissolution links in future networks, is important for mining and analyzing the evolution of social networks. In the past decade, many works have been done about the link prediction in social networks. The goal of this paper is to comprehensively review, analyze and discuss the state-of-the-art of the link prediction in social networks. A systematical category for link prediction techniques and problems is presented. Then link prediction techniques and problems are analyzed and discussed. Typical applications of link prediction are also addressed. Achievements and roadmaps of some active research groups are introduced. Finally, some future challenges of the link prediction in social networks are discussed.
Article
Recovering and reconstructing networks by accurately identifying missing and unreliable links is a vital task in the domain of network analysis and mining. In this article, by studying a specific local structure, namely, a degree block having a node and its all immediate neighbors, we find it contains important statistical features of link formation for complex networks. We therefore propose a parameter-free local blocking (LB) predictor to quantitatively detect link formation in given networks via local link density calculations. The promising experimental results performed on six real-world networks suggest that the new index can outperform other traditional local similarity-based methods on most of tested networks. After further analyzing the scores' correlations between LB and two other methods, we find that LB index simultaneously captures the features of both PA index and short-path-based index, which empirically verifies that LB index is a multiple-mechanism-driven link predictor.
Article
Community detection is an important task in network analysis, in which we aim to learn a network partition that groups together vertices with similar community-level connectivity patterns. By finding such groups of vertices with similar structural roles, we extract a compact representation of the network's large-scale structure, which can facilitate its scientific interpretation and the prediction of unknown or future interactions. Popular approaches, including the stochastic block model (SBM), assume edges are unweighted, which limits their utility by discarding potentially useful information. We introduce the weighted stochastic block model (WSBM), which generalizes the SBM to networks with edge weights drawn from any exponential family distribution. This model learns from both the presence and weight of edges, allowing it to discover structure that would otherwise be hidden when weights are discarded or thresholded. We describe a Bayesian variational algorithm for efficiently approximating this model's posterior distribution over latent block structures. We then evaluate the WSBM's performance on both edge-existence and edge-weight prediction tasks for a set of real-world weighted networks. In all cases, the WSBM performs as well or better than the best alternatives on these tasks.
Article
I investigate how social embeddedness affects an organization's acquisition and cost of financial capital in middle-market banking-a lucrative but understudied financial sector. Using existing theory and original fieldwork, I develop a framework to explain how embeddedness can influence which firms get capital and at what cost. I then statistically examine my claims using national data on small-business lending. At the level of dyadic ties, I find that firms that embed their commercial transactions with their lender in social attachments receive lower interest rates on loans. At the network level, firms are more likely to get loans and to receive lower interest rates on loans if their network of bank ties has a mix of embedded ties and arm's-length ties. These network effects arise because embedded ties motivate network partners to share private resources, while arm's-length ties facilitate access to public information on market prices and loan opportunities so that the benefits of different types of ties are optimized within one network. I conclude with a discussion of how the value produced by a network is at a premium when it creates a bridge that links the public information of markets with the private resources of relationships.
Article
Missing link prediction in networks is of both theoretical interest and practical significance in modern science. In this paper, we empirically investigate a simple framework of link prediction on the basis of node similarity. We compare nine well-known local similarity measures on six real networks. The results indicate that the simplest measure, namely Common Neighbours, has the best overall performance, and the Adamic-Adar index performs second best. A new similarity measure, motivated by the resource allocation process taking place on networks, is proposed and shown to have higher prediction accuracy than common neighbours. It is found that many links are assigned the same scores if only the information of the nearest neighbours is used. We therefore design another new measure exploiting information on the next nearest neighbours, which can remarkably enhance the prediction accuracy.
Article
Social support research has been hampered by a lack of clarity both in the definitions of social support and in the conceptualization of its effects on health outcomes. The present study compared social network size and three types of perceived social support—tangible, emotional, and informational —in relation to stressful life events, psychological symptoms and morale, and physical health status in a sample of 100 persons 45–64 years old. Social network size was empirically separable from, though correlated with, perceived social support and had a weaker overall relationship to outcomes than did support. Low tangible support and emotional support, in addition to certain life events, were independently related to depression and negative morale; informational support was associated with positive morale. Neither social support nor stressful life events were associated with physical health. It was concluded that social support research would benefit from attention to the multidimensionality of support and greater specificity in hypotheses about the relationship between types of support and adaptational outcomes.
Article
The Internet has become a rich and large repository of information about us as individuals. Anything from the links and text on a user’s homepage to the mailing lists the user subscribes to are reflections of social interactions a user has in the real world. In this paper we devise techniques and tools to mine this information in order to extract social networks and the exogenous factors underlying the networks’ structure. In an analysis of two data sets, from Stanford University and the Massachusetts Institute of Technology (MIT), we show that some factors are better indicators of social connections than others, and that these indicators vary between user populations. Our techniques provide potential applications in automatically inferring real world connections and discovering, labeling, and characterizing communities.
Conference Paper
Recommender systems can provide valuable services in a digital library environment, as demonstrated by its commercial success in book, movie, and music industries. One of the most commonly-used and successful recommendation algorithms is collaborative filtering, which explores the correlations within user-item interactions to infer user interests and preferences. However, the recommendation quality of collaborative filtering approaches is greatly limited by the data sparsity problem. To alleviate this problem we have previously proposed graph-based algorithms to explore transitive user-item associations. In this paper, we extend the idea of analyzing user-item interactions as graphs and employ link prediction approaches proposed in the recent network modeling literature for making collaborative filtering recommendations. We have adapted a wide range of linkage measures for making recommendations. Our preliminary experimental results based on a book recommendation dataset show that some of these measures achieved significantly better performance than standard collaborative filtering algorithms.
Article
Thesupport-vector network is a new learning machine for two-group classification problems. The machine conceptually implements the following idea: input vectors are non-linearly mapped to a very high-dimension feature space. In this feature space a linear decision surface is constructed. Special properties of the decision surface ensures high generalization ability of the learning machine. The idea behind the support-vector network was previously implemented for the restricted case where the training data can be separated without errors. We here extend this result to non-separable training data. High generalization ability of support-vector networks utilizing polynomial input transformations is demonstrated. We also compare the performance of the support-vector network to various classical learning algorithms that all took part in a benchmark study of Optical Character Recognition.
Article
Plenty of algorithms for link prediction have been proposed and were applied to various real networks. Among these algorithms, the weights of links are rarely taken into account. In this letter, we use local similarity indices to estimate the likelihood of the existence of links in weighted networks, including Common Neighbor, Adamic-Adar Index, Resource Allocation Index, and their weighted versions. We have tested the prediction accuracy on real social, technological and biological networks. Overall speaking, the resource allocation index performs best. To our surprise, sometimes the weighted indices perform even worse than the unweighted indices, which reminds us of the well-known Weak-Ties Theory. Further experimental study shows that the weak ties play a significant role in the link prediction, and to emphasize the contributions of weak ties can remarkably enhance the prediction accuracy for some networks. We give a semi-quantitative explanation based on the motif analysis. This letter provides a start point for the possible weak-ties theory in information retrieval.
Article
Network analysis is currently used in a myriad of contexts, from identifying potential drug targets to predicting the spread of epidemics and designing vaccination strategies and from finding friends to uncovering criminal activity. Despite the promise of the network approach, the reliability of network data is a source of great concern in all fields where complex networks are studied. Here, we present a general mathematical and computational framework to deal with the problem of data reliability in complex networks. In particular, we are able to reliably identify both missing and spurious interactions in noisy network observations. Remarkably, our approach also enables us to obtain, from those noisy observations, network reconstructions that yield estimates of the true network properties that are more accurate than those provided by the observations themselves. Our approach has the potential to guide experiments, to better characterize network data sets, and to drive new discoveries.
Article
We study empirically the time evolution of scientific collaboration networks in physics and biology. In these networks, two scientists are considered connected if they have coauthored one or more papers together. We show that the probability of a pair of scientists collaborating increases with the number of other collaborators they have in common, and that the probability of a particular scientist acquiring new collaborators increases with the number of his or her past collaborators. These results provide experimental evidence in favor of previously conjectured mechanisms for clustering and power-law degree distributions in networks.
Article
Networks have in recent years emerged as an invaluable tool for describing and quantifying complex systems in many branches of science. Recent studies suggest that networks often exhibit hierarchical organization, in which vertices divide into groups that further subdivide into groups of groups, and so forth over multiple scales. In many cases the groups are found to correspond to known functional units, such as ecological niches in food webs, modules in biochemical networks (protein interaction networks, metabolic networks or genetic regulatory networks) or communities in social networks. Here we present a general technique for inferring hierarchical structure from network data and show that the existence of hierarchy can simultaneously explain and quantitatively reproduce many commonly observed topological properties of networks, such as right-skewed degree distributions, high clustering coefficients and short path lengths. We further show that knowledge of hierarchical structure can be used to predict missing connections in partly known networks with high accuracy, and for more general network structures than competing techniques. Taken together, our results suggest that hierarchy is a central organizing principle of complex networks, capable of offering insight into many network phenomena.
Conference Paper
Question-Answering Bulletin Boards (QABB), such as Yahoo! Answers and Windows Live QnA, are gaining popularity recently. Communications on QABB connect users, and the overall connections can be regarded as a social network. If the evolution of social networks can be predicted, it is quite useful for encouraging communications among users. This paper describes an improved method for predicting links based on weighted proximity measures of social networks. The method is based on an assumption that proximities between nodes can be estimated better by using both graph proximity measures and the weights of existing links in a social network. In order to show the effectiveness of our method, the data of Yahoo! Chiebukuro (Japanese Yahoo! Answers) are used for our experiments. The results show that our method outperforms previous approaches, especially when target social networks are sufficiently dense.
Article
Given a snapshot of a social network, can we infer which new interactions among its members are likely to occur in the near future? We formalize this question as the link prediction problem, and develop approaches to link prediction based on measures for analyzing the "proximity" of nodes in a network. Experiments on large co-authorship networks suggest that information about future interactions can be extracted from network topology alone, and that fairly subtle measures for detecting node proximity can outperform more direct measures.
  • H A Taha
Taha, H. A. Operations Research. An Introduction. Eighth Edition. (Pearson Education Inc., 2007).
A Tutorial on Support Vector Regression, NeuroCOLT
  • A J Smola
  • B Schlkopf
A.J. Smola, and B. Schlkopf, (1998), A Tutorial on Support Vector Regression, NeuroCOLT, Technical Report NC-TR-98-030, Royal Holloway College, University of London, UK.
Supervised link prediction in weighted networks
  • De S H R
  • Prudłncio R B C
de S H R, Prudłncio R B C. Supervised link prediction in weighted networks. In: Proceedings of the 2011 International Joint Conference on Neural Networks (IJCNN), San Jose, 2011. 22812288
  • G H Golub
  • C F Van Loan
Golub, G. H. and Van Loan, C. F. Matrix Computations. 3rd edn (Baltimore MD: Johns Hopkins University Press, 1996).
prediction in complex networks: A local nave Bayes model
  • Z Liu
  • Q M Zhang
  • L L Zhou
  • T Link
The relationship between precision-recall and roc curves. Technical report 1551, University of Wisconsin Madison
  • M Davis
  • Goadrich
Toward link predictability of complex networks
L, L. et al. Toward link predictability of complex networks. Proc. Natl. Acad. Sci. USA 112, 23252330 (2015).