Social Influence Maximization Research Articles

The influence maximization algorithm for integrating attribute graph clustering and heterogeneous graph transformer

In social networks, maximizing influence is an important research direction. However, traditional influence maximization algorithms often overlook the attribute information of nodes and the heterogeneity of networks, leading to inefficiency and inaccuracy in the propagation process. To address this issue, this study first constructs a social network influence maximization propagation model, and then combines auto-encoder and graph convolutional autoencoder to extract social network user attributes. Finally, Transformer is used to learn the feature representation of social network nodes. Moreover, a heterogeneous graph neural network is introduced to combine with Transformer to further learn the feature representation of nodes, so as to design an influence maximization algorithm combining attributed graph clustering and heterogeneous graph Transformer. The results showed that the loss values of the fusion algorithm on different datasets were 0.619 and 0.17, respectively, proving its good fitting performance. The recall rate and F1 of this fusion algorithm were 92.5 % and 0.90, respectively, proving its high clustering accuracy. The influence maximization model based on fusion algorithm achieved active node coverage of 67 % and 48 % on two datasets, respectively, proving its good influence propagation effect. The above results demonstrate that the designed model can effectively spread influence. This helps to better understand and utilize the influence dissemination mechanisms in social networks, thereby promoting the development of social network research.

Combinatorial Stochastic-Greedy Bandit

We propose a novel combinatorial stochastic-greedy bandit (SGB) algorithm for combinatorial multi-armed bandit problems when no extra information other than the joint reward of the selected set of n arms at each time step t in [T] is observed. SGB adopts an optimized stochastic-explore-then-commit approach and is specifically designed for scenarios with a large set of base arms. Unlike existing methods that explore the entire set of unselected base arms during each selection step, our SGB algorithm samples only an optimized proportion of unselected arms and selects actions from this subset. We prove that our algorithm achieves a (1-1/e)-regret bound of O(n^(1/3) k^(2/3) T^(2/3) log(T)^(2/3)) for monotone stochastic submodular rewards, which outperforms the state-of-the-art in terms of the cardinality constraint k. Furthermore, we empirically evaluate the performance of our algorithm in the context of online constrained social influence maximization. Our results demonstrate that our proposed approach consistently outperforms the other algorithms, increasing the performance gap as k grows.

An interaction-aware approach for social influence maximization

An interaction-aware approach for social influence maximization

Alternate Solutions for Influence Maximization: Beyond theoretical approximation by the Genetic Algorithm Framework

As a special case of Social Influence Maximization (SIM) problems, the top k-influencers problem aims to identify the top k actors, called members of the seed set, in a network whose endorsement of a given action causes it to be chosen by the largest possible set of the network’s members, whose cardinality is called the influence spread (IS). Greedy framework algorithms like StaticGreedyCELF deliver a ground-truth solution with a proven approximation. This manuscript aims to investigate solutions for SIM with advantages comparable to or superior to those of CELF’s solutions, such as the lower propagation time and variance of the IS. Novel models to discover alternatives to the CELF’s solution in the forms of the Risk-Averse (RAIM), Risk-Least (RLIM), Time-Least (TLIM), and Risk-Blind Influence Maximization (RBIM) models among solutions non-dominated by the CELF’s solution are proposed. A Genetic Algorithm Framework for Influence Maximization (GAFIM) is outlined, the only framework in the literature capable of achieving near-optimal solutions for the RAIM, RLIM, TLIM, and RBIM models under the famous diffusion models. A novel parameter tuning strategy offers GAFIM a class of optimal states rather than a single ideal state, allowing better exploration in each proposed model under each diffusion model. Notable findings from the research summary include (a) uncertainty in discovering an alternative under all conditions; (b) the favorable effect of expanding the seed set; (c) challenges in dominating CELF with RBIM; (d) a homogeneous variance distribution with no greater influence depth among alternatives; and (e) the positive side effects of models.

Sequential seeding policy on social influence maximization: a Q-learning-driven discrete differential evolution optimization

Sequential seeding policy on social influence maximization: a Q-learning-driven discrete differential evolution optimization

Influence- and Interest-Based Worker Recruitment in Crowdsourcing Using Online Social Networks

Workers recruitment remains a significant issue in Mobile Crowdsourcing (MCS), where the aim is to recruit a group of workers that maximizes the expected Quality of Service (QoS). Current recruitment systems assume that a pre-defined pool of workers is available. However, this assumption is not always true, especially in cold-start situations, where a new MCS task has just been released. Additionally, studies show that up to 96% of the available candidates are usually not willing to perform the assigned tasks. To tackle these issues, recent works use Online Social Networks (OSNs) and Influence Maximization (IM) to advertise about the desired MCS tasks through influencers, aiming to build larger pools. However, these works suffer from several limitations, such as 1) the lack of group-based selection methods when choosing influencers, 2) the lack of a well-defined worker recruitment process following IM, 3) and the non-dynamicity of the recruitment process, where the workers who refuse to perform the task are not substituted. In this paper, an Influence-and Interest-based Worker Recruitment System (IIWRS), using OSNs, is proposed. The proposed system has two main components: 1) an MCS-, group-, and interest-based IM approach, using a Genetic Algorithm, to select a set of influencers from the network to advertise about the MCS tasks, and 2) a dynamic worker recruitment process which considers the social attributes of workers, and is able to substitute those who do not accept to perform the assigned tasks. Empirical studies are performed using real-life datasets, while comparing IIWRS with existing benchmarks.

Submodular maximization over data streams with differential privacy noise

In the big data era, data often comes in the form of streams and fast data stream analysis has recently attracted intensive research interest. Submodular optimization naturally appears in many streaming data applications such as social network influence maximization with the property of diminishing returns. However, in a practical setting, streaming data frequently comes with noises that are small but significant enough to impact the optimality of submodular optimization solutions. Following the framework of differential privacy (DP), this paper considers a streaming model with DP noise that is small by construction. Within this noisy streaming model, the paper strives to address the general problem of submodular maximization with a cardinality constraint. The main theoretical result we obtained is a streaming algorithm that is one-pass and has an approximation guarantee of 1(2+(1+1k)2)(1+1k)−δ for any δ>0. Finally, we implement the algorithm and evaluate it against several baseline methods. Numerical results support the practical performance of our algorithm across several real datasets.

Special Issue on Adversarial AI to IoT Security and Privacy Protection: Attacks and Defenses

The prosperity of social IoT data brings revolutionary changes to our daily lives and greatly increases the existing data volume. But IoT data are vulnerable due to security and privacy issues. Over the past few years, malicious adversaries exploited various vulnerabilities of AI algorithms and thus compromised the security of AI systems. For example, obfuscating malware code within benign programs or applications to fool the AI-based intrusion detection systems. Thus, applying adversarial AI is supposed to be one of the most useful methods to protect IoT data, including big data mining and analysis, information diffusion, sentiment analysis and opinion mining, social event detection, trend prediction and influence maximization. This special issue brings together leading researchers and developers presenting their latest research and 10 high-quality papers are selected. A summary of these accepted papers is outlined below. In the paper entitled ‘AWFC: Preventing Label Flipping Attacks towards Federated Learning for Intelligent IoT’ by Zhuo Lv et al., the authors are motivated to prevent label flipping poisoning attacks by observing the changes in model parameters that were trained by different single labels. They propose a novel detection method, called AWFC, that label flipping attacks are detected by identifying the differences of classes in the data. The weight assignments in a fully connected layer of the neural network model are used and the statistical algorithm is applied to find the malicious clients. The experiments are conducted on benchmark data, such as Fashion-MNIST and Intrusion Detection Evaluation Dataset (CIC-IDS2017), where results demonstrate that the method’s detection accuracy is better.

A Discrete Particle Swarm Optimization Algorithm Based on Neighbor Cognition to Solve the Problem of Social Influence Maximization

<p>In view of the problem that the estimation method of node influence in social network is not comprehen-sive and the Particle Swarm Optimization (PSO) algorithm is easy to fall into the local optimal and the lo-cal search ability is insufficient. In this paper, we proposed a Neighbor Cognitive Discrete Particle Swarm Optimization (NCDPSO) algorithm. Aiming at the problem of influence in social networks, a new node influence measure method is proposed, the three-degree theory is introduced to comprehensively estimate the influence of nodes. In order to improve the global search ability of the PSO, the “neighbor cognition” factor is proposed to enhance the breadth of learning; and the following bee strategy is introduced to pro-pose particle density and survivability to control the number of elite clones, so as to solve the problem of insufficient local search ability of the algorithm. Finally, the validity of the proposed algorithm is verified by testing on real data sets and comparing with other algorithms.</p> <p> </p>

GNPA: a hybrid model for social influence maximization in dynamic networks

GNPA: a hybrid model for social influence maximization in dynamic networks

Reliability Maximization in Uncertain Graphs

Network reliability measures the probability that a target node is reachable from a source node in an uncertain graph, i.e., a graph where every edge is associated with a probability of existence. In this paper, we investigate the novel and fundamental problem of adding a small number of edges in the uncertain network for maximizing the reliability between a given pair of nodes. We study the <inline-formula><tex-math notation="LaTeX">$\mathbf {NP}$</tex-math></inline-formula> -hardness and the approximation hardness of our problem, and design effective, scalable solutions. Furthermore, we consider extended versions of our problem (e.g., multiple source and target nodes can be provided as input) to support and demonstrate a wider family of queries and applications, including sensor network reliability maximization and social influence maximization. Experimental results validate the effectiveness and efficiency of the proposed algorithms.

Stochastic Top K -Subset Bandits with Linear Space and Non-Linear Feedback with Applications to Social Influence Maximization

There are numerous real-world problems where a user must make decisions under uncertainty. For the problem of influence maximization on a social network, for example, the user must select a set of K influencers who will jointly have a large influence on many users. With the lack of prior knowledge about the diffusion process or even topological information, this problem becomes quite challenging. This problem can be cast as a combinatorial bandit problem, where the user can repeatedly choose a candidate set of K out of N arms at each time, with an aim to achieve an efficient trade-off between exploration and exploitation. In this work, we present the first combinatorial bandit algorithm for which the only feedback is a non-linear reward of the selected K arms. No other feedback is needed. In the context of influence maximization, this means no feedback in the form of which nodes or edges were activated needs to be available, just the amount of influence. The novel algorithm we propose, CMAB-SM, is based on a divide-and-conquer strategy. It is computationally and storage efficient. Over a time horizon T , the proposed algorithm achieves a regret bound of Õ( K 1/2 N 1/3 T 2/3 ). This bound is sub-linear in all of the parameters: T , N , and K . We empirically demonstrate our algorithm’s performance using the applications of influence maximization and product cross-selling. For influence maximization, we provide experiments on real-world social networks, showing that the proposed CMAB algorithm outperforms bandit-specific and social-influence-domain-specific algorithms in terms of empirical run-time and expected influence. For product cross-selling, we also demonstrate that the proposed CMAB algorithm outperforms considered baselines on synthetic data.

A quick GRASP-based method for influence maximization in social networks

The evolution and spread of social networks have attracted the interest of the scientific community in the last few years. Specifically, several new interesting problems, which are hard to solve, have arisen in the context of viral marketing, disease analysis, and influence analysis, among others. Companies and researchers try to find the elements that maximize profit, stop pandemics, etc. This family of problems is collected under the term Social Network Influence Maximization problem (SNIMP), whose goal is to find the most influential users (commonly known as seeds) in a social network, simulating an influence diffusion model. SNIMP is known to be an mathcal {NP}-hard problem and, therefore, an exact algorithm is not suitable for solving it optimally in reasonable computing time. The main drawback of this optimization problem lies on the computational effort required to evaluate a solution. Since each node is infected with a certain probability, the objective function value must be calculated through a Monte Carlo simulation, resulting in a computationally complex process. The current proposal tries to overcome this limitation by considering a metaheuristic algorithm based on the Greedy Randomized Adaptive Search Procedure (GRASP) framework to design a quick solution procedure for the SNIMP. Our method consists of two distinct stages: construction and local search. The former is based on static features of the network, which notably increases its efficiency since it does not require to perform any simulation during construction. The latter involves a local search based on an intelligent neighborhood exploration strategy to find the most influential users based on swap moves, also aiming for an efficient processing. Experiments performed on 7 well-known social network datasets with 5 different seed set sizes confirm that the proposed algorithm is able to provide competitive results in terms of quality and computing time when comparing it with the best algorithms found in the state of the art.

GRAIN

Data selection methods, such as active learning and core-set selection, are useful tools for improving the data efficiency of deep learning models on large-scale datasets. However, recent deep learning models have moved forward from independent and identically distributed data to graph-structured data, such as social networks, e-commerce user-item graphs, and knowledge graphs. This evolution has led to the emergence of Graph Neural Networks (GNNs) that go beyond the models existing data selection methods are designed for. Therefore, we present GRAIN, an efficient framework that opens up a new perspective through connecting data selection in GNNs with social influence maximization. By exploiting the common patterns of GNNs, GRAIN introduces a novel feature propagation concept, a diversified influence maximization objective with novel influence and diversity functions, and a greedy algorithm with an approximation guarantee into a unified framework. Empirical studies on public datasets demonstrate that GRAIN significantly improves both the performance and efficiency of data selection (including active learning and core-set selection) for GNNs. To the best of our knowledge, this is the first attempt to bridge two largely parallel threads of research, data selection, and social influence maximization, in the setting of GNNs, paving new ways for improving data efficiency.

Product Adoption Maximization Leveraging Social Influence and User Interest Mining

A Social Networking Service (SNS) platform provides digital footprints to discover users’ interests and track the social diffusion of product adoptions. How to identify a small set of seed users in a SNS who is potential to adopt a new promoting product with high probability, is a key question in social networks. Existing works approached this as a social influence maximization problem. However, these approaches relied heavily on text information for topic modeling and neglected the impact of seed users’ relation in the model. To this end, in this paper, we first develop a general product adoption function integrating both users’ interest and social influence, where the user interest model relies on historical user behavior and the seed users’ evaluations without any text information. Accordingly, we formulate a product adoption maximization problem and prove NP-hardness of this problem. We then design an efficient algorithm to solve this problem. We further devise a method to automatically learn the parameter in the proposed adoption function from users’ past behaviors. Finally, experimental results show the soundness of our proposed adoption decision function and the effectiveness of the proposed seed selection method for product adoption maximization.

Social Influence Maximization in Hypergraphs.

This work deals with a generalization of the minimum Target Set Selection (TSS) problem, a key algorithmic question in information diffusion research due to its potential commercial value. Firstly proposed by Kempe et al., the TSS problem is based on a linear threshold diffusion model defined on an input graph with node thresholds, quantifying the hardness to influence each node. The goal is to find the smaller set of items that can influence the whole network according to the diffusion model defined. This study generalizes the TSS problem on networks characterized by many-to-many relationships modeled via hypergraphs. Specifically, we introduce a linear threshold diffusion process on such structures, which evolves as follows. Let be a hypergraph. At the beginning of the process, the nodes in a given set are influenced. Then, at each iteration, (i) the influenced hyperedges set is augmented by all edges having a sufficiently large number of influenced nodes; (ii) consequently, the set of influenced nodes is enlarged by all the nodes having a sufficiently large number of already influenced hyperedges. The process ends when no new nodes can be influenced. Exploiting this diffusion model, we define the minimum Target Set Selection problem on hypergraphs (TSSH). Being the problem NP-hard (as it generalizes the TSS problem), we introduce four heuristics and provide an extensive evaluation on real-world networks.

Social network node influence maximization method combined with degree discount and local node optimization

For the problem of maximizing the influence of nodes in traditional social networks, it is impossible to select nodes and spread the scope of large-scale diffusion simultaneously. Based on local node optimization and degree discount, a new node influence maximization algorithm is proposed (degree discount and local improvement method, DLIM). Firstly, the candidate seed set is optimized. The NAV (Node Approximate Influence Value) function is constructed to calculate the Influence Value of local Node. Determine nodes with similar influence value and select the source node; the similarity method is used to filter and delete nodes with similar influence value. Secondly, a node activation algorithm is proposed to filter candidate nodes with the idea of degree discount. DMAP (degree discount and maximum activation probability) is constructed. The function uses the filtered candidate nodes for global diffusion. Finally, the proposed DLIM algorithm is used to select the seed nodes to optimize the nodes, and the independent cascade propagation model is used to conduct comparative analysis experiments on four real data sets of Wiki-Vote, NetHEPT, NetPHY, and GrQc four real datasets, and with the Independent Cascade (IC). A comparative analysis experiment is carried out on the propagation model. The experimental results show that: the proposed DLIM algorithm improves the propagation range by 11.3% compared with the traditional degree discount algorithm. The time efficiency is four orders of magnitude faster than the traditional degree discount algorithm. The proposed DLIM algorithm is reasonable and effective. It can also be applied in network marketing, product recommendation, and other fields.

A Node Embedding-Based Influential Spreaders Identification Approach

Node embedding is a representation learning technique that maps network nodes into lower-dimensional vector space. Embedding nodes into vector space can benefit network analysis tasks, such as community detection, link prediction, and influential node identification, in both calculation and richer application scope. In this paper, we propose a two-step node embedding-based solution for the social influence maximization problem (IMP). The solution employs a revised network-embedding algorithm to map input nodes into vector space in the first step. In the second step, the solution clusters the vector space nodes into subgroups and chooses the subgroups’ centers to be the influential spreaders. The proposed approach is a simple but effective IMP solution because it takes both the social reinforcement and homophily characteristics of the social network into consideration in node embedding and seed spreaders selection operation separately. The information propagation simulation experiment of single-point contact susceptible-infected-recovered (SIR) and full-contact SIR models on six different types of real network data sets proved that the proposed social influence maximization (SIM) solution exhibits significant propagation capability.

Approximating special social influence maximization problems

Social Influence Maximization Problems (SIMPs) deal with selecting k seeds in a given Online Social Network (OSN) to maximize the number of eventually-influenced users. This is done by using these seeds based on a given set of influence probabilities among neighbors in the OSN. Although the SIMP has been proved to be NP-hard, it has both submodular (with a natural diminishing-return) and monotone (with an increasing influenced users through propagation) that make the problem suitable for approximation solutions. However, several special SIMPs cannot be modeled as submodular or monotone functions. In this paper, we look at several conditions under which non-submodular or non-monotone functions can be handled or approximated. One is a profit-maximization SIMP where seed selection cost is included in the overall utility function, breaking the monotone property. The other is a crowd-influence SIMP where crowd influence exists in addition to individual influence, breaking the submodular property. We then review several new techniques and notions, including double-greedy algorithms and the supermodular degree, that can be used to address special SIMPs. Our main results show that for a specific SIMP model, special network structures of OSNs can help reduce its time complexity of the SIMP.

A survey on influence maximization in a social network

Given a social network with diffusion probabilities as edge weights and a positive integer k, which k nodes should be chosen for initial injection of information to maximize the influence in the network? This problem is popularly known as the Social Influence Maximization Problem (SIM Problem). This is an active area of research in computational social network analysis domain, since one and half decades or so. Due to its practical importance in various domains, such as viral marketing, target advertisement and personalized recommendation, the problem has been studied in different variants, and different solution methodologies have been proposed over the years. This paper presents a survey on the progress in and around SIM Problem. At last, it discusses current research trends and future research directions as well.