Influential Nodes In Social Networks Research Articles

A Graph Sampling-Based Model for Influence Maximization in Large-Scale Social Networks

Social networks have attracted a great deal of attention and have, in fact, changed the way we produce, consume, and diffuse information. This change gave rise to the notion of social influence and today we talk about influential nodes. The process of detecting influential nodes in social networks aims to find entities that propagate information to a large portion of the network users. This process is often known as the influence maximization (IM) problem. Due to the explosive growth of social networks’ data, their structure is more complex and we talk about “big graph data.” Moreover, modern networks are dynamic and their topology or/and information is likely to change over time. Detecting influential nodes in such networks is a challenging task. Several methods have been developed in this context. However, they concentrate on static networks and there is little work on large-scale social networks. We propose in this article a new model for IM called MapReduce-based dynamic selection of influential nodes (MR-DSINs) that has the ability to cope with the huge size of real social networks. In fact, our approach is based on a graph sampling step in order to reduce the network’s size. Given that reduced version, MR-DSIN is able to select dynamically influential nodes. Our proposal has the advantage of considering the dynamics of information that can be modeled by users’ social actions (e.g.“, share”, “comment”, “retweet”). Experimental results on real-world social networks and computer-generated artificial graphs demonstrate that MR-DSIN is efficient for identifying influential nodes, compared with three known proposals. We prove that our model is able to detect in the reduced graph an influence as important as in the original one.

Multi-factor information matrix: A directed weighted method to identify influential nodes in social networks

Identifying super-propagators on social networks takes on critical significance in guiding and controlling public opinions on the internet, information retrieval, viral marketing and so on. The influence of nodes is enhanced under the dual role of network structure and dynamic interactions, while existing research tends to attach great importance on the network structure. This means the interaction behaviors between nodes are ignored, which indicates that the performance of algorithms still needs to be improved. Moreover, only a minority of work has been conducted on directed weighted networks. To address these problems, from the perspectives of the network topology and dynamic interaction, a multi-factor information matrix centrality algorithm is developed in this study in accordance with the characteristics of information propagation process and three degrees of influence rule, with the node influence, neighbor influence, and mutual influence exerted by information feedback considered simultaneously. Based on this, the subjective and objective weighted methods are combined to enhance the accuracy of evaluation results, then the multi-factor matrix can be obtained. Four comparative experiments are employed to verify the effectiveness of proposed method with ten networks and nine existing centralities applied. The results of the experiments confirm the validity and superiority of the proposed method in terms of differentiation, connectivity and propagation performance. Moreover, its low time complexity makes the algorithm can be used in large-scale social networks. In general, methodological and multidimensional support is provided for the accurate identification of super-propagators in online social networks.

PHEE: Identifying influential nodes in social networks with a phased evaluation-enhanced search

How to identify a small fraction of users with the best capacity to influence other users is central to the research of social network analysis. This problem is termed influence maximization (IM) and is known for its extensive applications. IM can be formulated as a combinatorial optimization problem, which has been shown to be NP-hard under some diffusion models. Therefore, seeking for high-accuracy IM algorithms with acceptable running time has attracted much attention in literature. However, most of existing IM algorithms only adopt a uniform mechanism in the whole solution search process, which lacks of flexible response when the algorithms trap in local optimum. This paper proposes a phased evaluation-enhanced (PHEE) approach for IM, which utilizes two distinct strategies to search the optimal solutions: The first one is a random range division-based evolutionary algorithm for improving solution quality; the second is a fast convergence strategy for searching an optimal solution. Two PHEE-based algorithms, MDD-PHEE and GCI-PHEE, are generated and are evaluated on 10 real-world social networks of different sizes and types. Experimental results demonstrate the effectiveness of PHEE; in particular, MDD-PHEE obtains the best influence spread on all networks compared with the state-of-the-art algorithms, and has a better performance than the time-consuming algorithm CELF on four datasets.

Identification of Influential Nodes in Social Network: Graph Approach

The identification of nodes that are actually influencing other nodes is one of the core and fundamental research problems in the social network data sciences. There are multiple proven techniques available globally that depict the influence of the root node on the remaining nodes in the given network. Although methods and methodology is already proven and multiple experiments have been carried out in this area, the current research work is an extension of the influential node identification problem with the perspective of graph traversal. To prove the methodology and new approach a tree and AUC (area under the curve) - based analysis method has been proposed, TArank in which the information collected from the breadth-first search tree is used to identify influential nodes. From the obtained results and analysis, the effectiveness of TARank in the identification of influential nodes has been validated. The concentration of data at a particular segment of the plot is observed. Additionally, the tree view of the analysis predicts the root and leaf node thus helping in the identification of the most influential nodes in the online social media network. Keywords: Influential Node, PCC, TARank, Degree Centrality, PCC

Branching processes reveal influential nodes in social networks

Branching processes are discrete-time stochastic processes which have been largely employed to model and simulate information diffusion processes over large online social networks such as Twitter and Reddit. Here we show that a variant of the branching process model enables the prediction of the popularity of user-generated content and thus can serve as a method for ranking search results or suggestions displayed to users. The proposed branching-process variant is able to evaluate the importance of an agent in a social network and, thus we propose a novel centrality index, called the Stochastic Potential Gain (SPG). The SPG is the first centrality index which combines the knowledge of the network topology with a dynamic process taking place on it which we call a graph-driven branching process. SPG generalises a range of popular network centrality metrics such as Katz' and Subgraph. We formulate a Monte Carlo algorithm (called MCPG) to compute the SPG and prove that it is convergent and correct. Experiments on two real datasets drawn from Facebook and GitHub demonstrate that MCPG traverses only a small fraction of nodes to produce its result, thus making the Stochastic Potential Gain an appealing option to compute node centrality measure for Online social networks.

Identifying Influential Nodes in Social Networks: Exploiting Self-Voting Mechanism.

The influence maximization (IM) problem is defined as identifying a group of influential nodes in a network such that these nodes can affect as many nodes as possible. Due to its great significance in viral marketing, disease control, social recommendation, and so on, considerable efforts have been devoted to the development of methods to solve the IM problem. In the literature, VoteRank and its improved algorithms have been proposed to select influential nodes based on voting approaches. However, in the voting process of these algorithms, a node cannot vote for itself. We argue that this voting schema runs counter to many real scenarios. To address this issue, we designed the VoteRank* algorithm, in which we first introduce the self-voting mechanism into the voting process. In addition, we also take into consideration the diversities of nodes. More explicitly, we measure the voting ability of nodes and the amount of a node voting for its neighbors based on the H-index of nodes. The effectiveness of the proposed algorithm is experimentally verified on 12 benchmark networks. The results demonstrate that VoteRank* is superior to the baseline methods in most cases.

Privacy-preserving identification of the influential nodes in networks

Identifying influential nodes in social networks has drawn significant attention in the field of network science. However, most of the existing works request to know the complete structural information about networks, indeed, this information is usually sensitive, private and hard to obtain. Therefore, how to identify the influential nodes in networks without disclosing privacy is especially important. In this paper, we propose a privacy-preserving (named as HE-ranking) framework to identify influential nodes in networks based on homomorphic encryption (HE) protocol. The HE-ranking method collaboratively computes the nodes’ importance and protects the sensitive information of each private network by using the HE protocol. Extensive experimental results indicate that the method can effectively identify the influential nodes in the original networks than the baseline methods which only use each private network to identify influential nodes. More importantly, the HE-ranking method can protect the privacy of each private network in different parts.

Identify influential nodes in social networks with graph multi-head attention regression model

Identifying influential nodes in social networks is a fundamental task. Due to the development of Graph Neural Networks, Graph Convolution Network (GCN) based model has been introduced to solve this problem. Compared to traditional methods, the existing GCN-based models are more accurate in identifying influential nodes because they can better aggregate the multi-dimension features. However, the GCN-based method treats this problem as a binary classification task rather than a regression task, making it less practical. To make the GCN-based model more practical, we treat identifying influential nodes as a regression task. Moreover, when aggregating neighbor features, GCN ignores the difference in neighbor importance, which will affect the prediction performance of the GCN-based models. This paper proposes a graph multi-head attention regression model to address these problems. Vast experiments on twelve real-world social networks demonstrate that the proposed model significantly outperforms baseline methods. To the best of our knowledge, this is the first work to introduce the multi-head attention mechanism to identify influential nodes in social networks.

MCD: A modified community diversity approach for detecting influential nodes in social networks

Over the last couple of decades, Social Networks have connected people on the web from across the globe and have become a crucial part of our daily life. These networks have also rapidly grown as platforms for propagating products, ideas, and opinions to target a wider audience. This calls for the need to find influential nodes in a network for a variety of reasons, including the curb of misinformation being spread across the networks, advertising products efficiently, finding prominent protein structures in biological networks, etc. In this paper, we propose Modified Community Diversity (MCD), a novel method for finding influential nodes in a network by exploiting community detection and a modified community diversity approach. We extend the concept of community diversity to a two-hop scenario. This helps us evaluate a node’s possible influence over a network more accurately and also avoids the selection of seed nodes with an overlapping scope of influence. Experimental results verify that MCD outperforms various other state-of-the-art approaches on eight datasets cumulatively across three performance metrics.

Influence maximization in social networks using transfer learning via graph-based LSTM

Social networks have emerged as efficient platforms to connect people worldwide and facilitate the rapid spread of information. Identifying influential nodes in social networks to accelerate the spread of particular information across the network formulates the study of influence maximization (IM). In this paper, inspired by deep learning techniques, we propose a novel approach to solve the influence maximization problem as a classical regression task using transfer learning via graph-based long short-term memory (GLSTM). We start by calculating three popular node centrality methods as feature vectors for the nodes in the network and every node’s individual influence under susceptible–infected–recovered (SIR) information diffusion model, which forms the labels of the nodes in the network. The generated feature vectors and their corresponding labels for all the nodes are then fed into a graph-based long short-term memory (GLSTM). The proposed architecture is trained on a vast and complex network to generalize the model parameters better. The trained model is then used to predict the probable influence of every node in the target network. The proposed model is compared with some of the well-known and recently proposed algorithms of influence maximization on several real-life networks using the popular SIR model of information diffusion. The intensive experiments suggest that the proposed model outperforms these well-known and recently proposed influence maximization algorithms.

Identifying influential nodes in social networks: Centripetal centrality and seed exclusion approach

Identifying influential nodes in a network is vital for the study of social network structure and to facilitate the dissemination of requisite information. The challenge we address is that, given a complex network, which nodes are more important? How can a group of disseminators be identified and selected to maximize any given field of influence? A series of centrality measures are proposed from different perspectives based on the topology of nodes. However existing methods suffer from problems that are intrinsic to singular consideration of node topology information, and they neglect the connection relationship between nodes when filtering the spreaders, resulting in imprecise evaluation results and limited spread scale. To solve this issue, this paper proposes a new centrality, inspired by the centripetal force formula. Centripetal centrality combines global, and local, as well as semi-local topological information about the nodes resulting in a more comprehensive evaluation. For the problems related to influence maximization, we propose a heuristic algorithm called seed exclusion (SE) that filters propagators. To demonstrate the performance of the proposed measures, we conducted experiments on both real-world and synthetic networks by comparing distinct metrics, improvements in network efficiency, the propagation of nodes under the SIR model and the average shortest distance between spreaders. The experimental results show that the proposed centripetal centrality is more accurate and effective than similar measures, while comparison with baselines the SE algorithm significantly improves spread speed and infection scale.

Influence maximization problem by leveraging the local traveling and node labeling method for discovering most influential nodes in social networks

The influence maximization problem has gained particular importance in viral marketing for large-scale spreading in social networks. Developing a fast and appropriate algorithm to identify an optimized seed set for the diffusion process on social networks is crucial due to the fast growth of networks. Most fast methods only focus on the degree of nodes while ignoring the strategic position of nodes in the networks. These methods do not have the required quality in finding a seed set in most networks. On the other hand, many other methods have acceptable quality, but their computational overhead is significant. To address these issues, the main concentration of this paper is to propose a fast and accurate method for the influence maximization problem, which uses a local traveling for labeling of nodes based on the influence power, called the LMP algorithm. In the proposed LMP algorithm, first, a travel starts from a node with the lowest influence power to assign a ranking-label for this node and its neighbor nodes in each step based on their diffusion capability and strategic position. The LMP algorithm uses node labeling steps to reduce search space significantly. Three ranking-labels are used in the proposed algorithm, and nodes with the highest ranking-label are selected as candidate nodes. This local and fast step strictly reduces the search space. Finally, the LMP algorithm selects seed nodes based on the topology features and the strategic position of the candidate and connector. The performance of the proposed algorithm is benchmarked with the well-known and recently proposed seed selection algorithms. The experimental results are performed on real-world and synthetic networks to validate the efficiency and effectiveness. The experiments exhibit that the proposed algorithm is the fastest in comparison with other state-of-the-art algorithms, and it has linear time complexity. In addition, it can achieve a good tradeoff between the efficiency and time complexity in the influence maximization problem.

Identifying influential nodes in social networks: A voting approach

With the prosperity of social networks, the research of influence maximization has become growing importance and captured increasing attention from various disciplines. The key point in influence maximization is to identify a group of influential nodes that are scattered broadly in a network. In this regard, we propose the VoteRank++ method, which is a voting approach, to iteratively select the influential nodes. In the viewpoint of VoteRank++, nodes with different degrees should carry different amounts of votes in consideration of the diversity of nodes in voting ability, and a node may vote differently for its neighbors by considering the varying degrees of closeness between nodes. Moreover, to reduce the overlapping of influential regions of spreaders, VoteRank++ discounts the voting ability of 2-hop neighbors of the selected nodes. Then, to avoid the cost of calculating the voting scores of all nodes in each iteration, only the nodes whose scores may change need to update their voting scores. To demonstrate the effectiveness of the proposed method, we employ both the Susceptible-Infected-Recovered and Linear Threshold models to simulate the spreading progress. Experimental results show that VoteRank++ outperforms the baselines on both spreading speed and infected scale in most of the cases.

NodeRank: Finding influential nodes in social networks based on interests

NodeRank: Finding influential nodes in social networks based on interests

Positive opinion maximization in signed social networks

Opinion maximization is a kind of optimization method, which leverages a subset of influential nodes in social networks to spread user opinions towards the target product and eventually obtains the largest opinion propagation. The current propagation models on the opinion maximization mainly focus on the activated nodes and the static opinion formation process. However, they neglect the combination between the activated nodes and the dynamic opinion formation process. Moreover, previous studies are more attentive to the positive relationships among users. In the real scenario, negative relationships among users may damage the product reputation. Therefore, in this paper, we study positive opinion maximization by using an Activated Opinion Maximization Framework (AOMF) in signed social networks. The proposed AOMF is composed of three phases: i) the selection of candidate seed nodes, ii) the activated opinion formation process and iii) the determination of seed nodes. We first use an effective heuristic rule to select candidate seed nodes. To model the activation and dynamic opinion formation process of network nodes, we devise the activated opinion formation model based on the multi-stage linear threshold model and the Degroot model. Then, we calculate the opinion propagation of each candidate seed node by using the activated opinion formation model. Based on the candidate seed nodes and the activated opinion formation process, seed nodes are further determined. Finally, experimental results on six social network datasets demonstrate that the proposed method has superior potential opinions and positive ratio than the chosen benchmarks.

A dynamic algorithm based on cohesive entropy for influence maximization in social networks

The problem of influence maximization in social networks has been widely investigated, but most previous studies have usually ignored the dynamic nature of propagation and the effects of local aggregation factors on diffusion. This paper presents a Dynamic algorithm based on cohesive Entropy for Influence Maximization (DEIM), the goal of which is to find the most influential nodes in social networks. Firstly, the Community Overlap Propagation Algorithm based on Cohesive Entropy (CECOPA) is put forward for the discovery of overlapping communities in networks, and potential nodes in the gathering area are selected to construct the candidate seed set. Then, the Optional Dynamic influence Propagation algorithm (ODP) is designed based on narrowing down the selection range of seeds. It utilizes a variety of entropy calculations to obtain the cohesive power between neighboring nodes and then determines whether the node has the ability to become a propagable pioneer of another node; thus, information continues to diffuse effectively. Finally, via many times experiments on several data sets, it is confirmed that the proposed DEIM algorithm in this paper can successfully affect the ideal number of users in different scenarios.

Recognizing Influential Nodes in Social Networks With Controllability and Observability

The analysis for social networks, such as the sensor-networks in socially networked industries, has shown a deep influence of intelligent information processing technology on industrial systems. The large amounts of data on these networks raise the urgent demands of analyzing the topological content effectively and efficiently in Industrial Internet of Things. One of the ways to locate important information amongst such large troves of data is to recognize influential nodes. In this article, we examine an intelligent way to recognize the influence of such nodes automatically. Motivated by the concepts of system controllability and observability from control theory, we introduce a novel method to evaluate nodes from two different aspects, namely, the ability of “observe” information on the network (i.e., observability), and the ability to propagate information to other nodes (i.e., controllability). We propose a unified data mining framework that incorporates content analysis with nodes behavioral tendencies, and show that it is able to outperform competitive baselines in recognizing influential nodes in networks. We also show that it is important to detect the presence of spammer nodes within networks, which might otherwise be wrongly recognized as influential nodes. The experimental results demonstrate the superiority of the proposed approach in comparison with baseline methods.

Identifying Influential Nodes in Social Networks Based on Social Strength

Identifying influential nodes in social networks is significant in controlling the outbreak of epidemics, conducting advertisements for e-commercial products, predicting popular scientists or papers, and so on. Many methods have been proposed by paying much more attention to the role of nodes, while the social strength between nodes is rarely considered. Especially in unweighted networks, social strength is not explicit information. In this paper, we first characterize the social strength of unweighted networks, and then proposed a LOVital method based on the social strength. Numerical results on the relation network of Weibo dataset and the other datasets from various fields show that LOVital can much more accurately identify influential users.

Identifying influential nodes in social networks via community structure and influence distribution difference

This paper aims to effectively solve the problem of the influence maximization in social networks. For this purpose, an influence maximization method that can identify influential nodes via the community structure and the influence distribution difference is proposed. Firstly, the network embedding-based community detection approach is developed, by which the social network is divided into several high-quality communities. Secondly, the solution of influence maximization is composed of the candidate stage and the greedy stage. The candidate stage is to select candidate nodes from the interior and the boundary of each community using a heuristic algorithm, and the greedy stage is to determine seed nodes with the largest marginal influence increment from the candidate set through the sub-modular property-based Greedy algorithm. Finally, experimental results demonstrate the superiority of the proposed method compared with existing methods, from which one can further find that our work can achieve a good tradeoff between the influence spread and the running time.

Finding influential nodes in social networks based on neighborhood correlation coefficient

Finding the most influential nodes in social networks has significant applications. A number of methods have been recently proposed to estimate influentiality of nodes based on their structural location in the network. It has been shown that the number of neighbors shared by a node and its neighbors accounts for determining its influence. In this paper, an improved cluster rank approach is presented that takes into account common hierarchy of nodes and their neighborhood set. A number of experiments are conducted on synthetic and real networks to reveal effectiveness of the proposed ranking approach. We also consider ground-truth influence ranking based on Susceptible–Infected–Recovered model, on which performance of the proposed ranking algorithm is verified. The experiments show that the proposed method outperforms state-of-the-art algorithms.