With the rapid development and popularization of smart mobile devices, users tend to share their visited points-of-interest (POIs) on the network with attached location information, which forms a location-based social network (LBSN). LBSNs contain a wealth of valuable information, including the geographical coordinates of POIs and the social connections among users. Nowadays, lots of trust-enhanced approaches have fused the trust relationships of users together with other auxiliary information to provide more accurate recommendations. However, in the traditional trust-aware approaches, the embedding processes of the information on different graphs with different properties (e.g., user-user graph is an isomorphic graph, user-POI graph is a heterogeneous graph) are independent of each other and different embedding information is directly fused together without guidance, which limits their performance. More effective information fusion strategies are needed to improve the performance of trust-enhanced recommendation. To this end, we propose a T rust E nhanced POI recommendation approach with C ollaborative L earning (TECL) to merge geographic information and social influence. Our proposed model integrates two modules, a GAT-based graph autoencoder as trust relationships embedding module and a multi-layer deep neural network as a user-POI graph learning module. By applying collaborative learning strategy, these two modules can interact with each other. The trust embedding module can guide the selection of user’s potential features, and in turn the user-POI graph learning module enhances the embedding process of trust relationships. Different information is fused through the two-way interaction of information, instead of travelling in one direction. Extensive experiments are conducted using real-world datasets, and results illustrate that our suggested approach outperforms state-of-the-art methods.

Commit classification is an important task in software maintenance, since it helps software developers classify code changes into different types according to their nature and purpose. This allows them to better understand how their development efforts are progressing, identify areas where they need improvement, and make informed decisions about when and how to release new versions of their software. However, existing methods are all discriminative models, usually with complex architectures that require additional output layers to produce class label probabilities, making them task‐specific and unable to learn features across different tasks. Moreover, they require a large amount of labeled data for fine tuning, and it is difficult to learn effective classification boundaries in the case of limited labeled data. To solve the above problems, we propose a generative framework that incorporates prompt tuning for commit classification with external knowledge (IPCK), which simplifies the model structure and learns features across different tasks, only based on the commit message information as the input. First, we proposed a generative framework based on T5 (text‐to‐text transfer transformer). This encoder–decoder construction method unifies different commit classification tasks into a text‐to‐text problem, simplifying the model’s structure by not requiring an extra output layer. Second, instead of fine tuning, we design a prompt tuning solution that can be adopted in few‐shot scenarios with only limited samples. Furthermore, we incorporate external knowledge via an external knowledge graph to map the probabilities of words into the final labels in the speech machine step to improve performance in few‐shot scenarios. Extensive experiments on two open available datasets demonstrate that our framework can solve the commit classification problem simply but effectively for both single‐label binary classification and single‐label multiclass classification purposes with 90% and 83% accuracy. Further, in the few‐shot scenarios, our method improves the adaptability of the model without requiring a large number of training samples for fine tuning.

Influence maximization (IM) aims to find a group of influential nodes as initial spreaders to maximize the influence spread over a network. Yet, traditional IM algorithms have not been designed with fairness in mind, resulting in discrimination against some groups, like LGBTQ communities and racial minorities. This issue has spurred research on Fair Influence Maximization (FIM). However, existing FIM studies come with some drawbacks. First, most proposed notions of fairness for FIM cannot adjust the tradeoff between fairness level and influence spread. Second, though a few specific notions of fairness allow such balancing, they are limited to a few specific concave functions, which may not be suitable for various real-world scenarios. Furthermore, none of them have studied the deep relations between the features of concave functions and the level of fairness. Third, existing fairness metrics are limited to their corresponding concepts of fairness. Comparing the level of fairness across different algorithms using existing metrics can be challenging. To tackle the above problems, this article first proposes a novel fairness notion named Poverty Reward (PR), which achieves fairness by rewarding the enrichment of groups with low utility. Based on PR, we further propose an algorithmic framework called Concave Fairness Framework (CFF) that allows any concave function that satisfies specific requirements. We also systematically clarify how fairness is improved by applying concave functions and provide an in-depth quantitative analysis of how to select appropriate concave functions for different utility distributions. Moreover, we propose the Reward of Fairness (RoF) metric that evaluates the disparity between groups. Based on RoF, an evaluation system is built to uniformly compare FIM algorithms from different fairness notions. Experiments in real-world datasets have demonstrated the validity of the CFF, as well as the proposed fairness notion.

This paper studies the fair influence maximization problem with efficient algorithms. In particular, given a graph G, a community structure C consisting of disjoint communities, and a budget k, the problem asks to select a seed set S (|S| = k) that maximizes the influence spread while narrowing the influence gap between different communities. This problem derives from some significant social scenarios, such as health interventions (e.g. suicide/HIV prevention) where individuals from racial minorities or LGBTQ communities may be disproportionately excluded from the benefits of the intervention. To depict the concept of fairness in the context of influence maximization, researchers have proposed various notions of fairness, where the welfare fairness notion that better balances fairness level and influence spread has shown promising effectiveness. However, the lack of efficient algorithms for optimizing the objective function under welfare fairness restricts its application to networks of only a few hundred nodes. In this paper, we modify the objective function of welfare fairness to maximize the exponentially weighted sum and the logarithmically weighted sum over all communities' influenced fractions (utility). To achieve efficient algorithms with theoretical guarantees, we first introduce two unbiased estimators: one for the fractional power of the arithmetic mean and the other for the logarithm of the arithmetic mean. Then, by adapting the Reverse Influence Sampling (RIS) approach, we convert the optimization problem to a weighted maximum coverage problem. We also analyze the number of reverse reachable sets needed to approximate the fair influence at a high probability. Finally, we present an efficient algorithm that guarantees 1 − 1/e − ϵ (positive objective function) or 1 + 1/e + ϵ (negative objective function) approximation for any small ϵ > 0. Experiments demonstrate that our proposed algorithm could efficiently handle large-scale networks with good performance.

As one of the popular topics in social network analysis, link prediction aims to predict the likely but unobserved links between two nodes. It has a wide field of applications such as knowledge graph completion and recommender systems. Currently, graph neural network (GNN) is the most common method with excellent performance, which contains two main steps, i.e., node aggregation and edge concatenation. However, there are some drawbacks with the existing methods. Firstly, traditional node aggregation methods usually iteratively aggregate all or a fixed number of neighbors, which is inflexible and inefficient. Secondly, most of the existing edge concatenation methods only apply a single concatenation approach to obtain edge embeddings, which cannot fully guarantee the embedding quality. To tackle the two problems, this paper proposes a graph embedding approach for link prediction via decay coefficient based Proportional Aggregation and Hybrid Concatenation (PAHC). On one hand, PAHC directly aggregates only proportional neighbors from different orders by correlating the decaying phenomenon of information propagation in social networks with the decaying phenomenon of temperature in Newton’s cooling theorem. On the other hand, PAHC proposes a hybrid concatenation approach to obtain the final edge embedding by mixing weighted summation and weighted direct concatenation. Experiments on five datasets show that through decay proportional aggregation and hybrid concatenation, the proposed PAHC can better predict links in social networks, outperforming state-of-the-art methods.

Link prediction is an important task in social network analysis and mining because of its various applications. A large number of link prediction methods have been proposed. Among them, the deep learning-based embedding methods exhibit excellent performance, which encodes each node and edge as an embedding vector, enabling easy integration with traditional machine learning algorithms. However, there still remain some unsolved problems for this kind of methods, especially in the steps of node embedding and edge embedding. First, they either share exactly the same weight among all neighbors or assign a completely different weight to each node to obtain the node embedding. Second, they can hardly keep the symmetry of edge embeddings obtained from node representations by direct concatenation or other binary operations such as averaging and Hadamard product. In order to solve these problems, we propose a weighted symmetric graph embedding approach for link prediction. In node embedding, the proposed approach aggregates neighbors in different orders with different aggregating weights. In edge embedding, the proposed approach bidirectionally concatenates node pairs both forwardly and backwardly to guarantee the symmetry of edge representations while preserving local structural information. The experimental results show that our proposed approach can better predict network links, outperforming the state-of-the-art methods. The appropriate aggregating weight assignment and the bidirectional concatenation enable us to learn more accurate and symmetric edge representations for link prediction.

Graph Neural Networks (GNNs) have achieved great success in graph representation learning. Modern GNNs are built upon the homophily assumption and iteratively aggregate messages from immediate neighbors through the message-passing mechanism, which limits the ability of GNNs to represent graphs with heterophily. Existing GNNs considering heterophily adapt to the heterophilous graphs through reconstructing neighborhood and message fusion. However, noise information is constantly superimposed during message passing, due to the mixed propagation of different-order messages. In this paper, we propose a special graph convolutional network with virtual edge message passing (VEGCN), which consists of three important components: virtual edge message passing, message attention, and residual connection. Different from the message-passing mechanism of existing GNNs, virtual edge message passing can directly transfer messages from the second-order neighbors to the target nodes. By bypassing the first-order neighbors, VEGCN can avoid interference from first-order neighbors. In addition, we design an attention mechanism to adaptively obtain messages from first-order and second-order neighbors. This attention mechanism can distinguish the different importance of the messages from first-order neighbors and second-order neighbors. Finally, we introduce a residual connection to enhance the features of the nodes themselves and alleviate over-smoothing. We validate the effectiveness of VEGCN on several benchmark datasets including graphs with homophily and heterophily. Experimental results show that VEGCN outperforms representative baselines. Furthermore, we also designed ablation experiments to verify the role of the core components.

Given a graph $G$, a community structure $\mathcal{C}$, and a budget $k$, the fair influence maximization problem aims to select a seed set $S$ ($|S|\leq k$) that maximizes the influence spread while narrowing the influence gap between different communities. While various fairness notions exist, the welfare fairness notion, which balances fairness level and influence spread, has shown promising effectiveness. However, the lack of efficient algorithms for optimizing the welfare fairness objective function restricts its application to small-scale networks with only a few hundred nodes. In this paper, we adopt the objective function of welfare fairness to maximize the exponentially weighted summation over the influenced fraction of all communities. We first introduce an unbiased estimator for the fractional power of the arithmetic mean. Then, by adapting the reverse influence sampling (RIS) approach, we convert the optimization problem to a weighted maximum coverage problem. We also analyze the number of reverse reachable sets needed to approximate the fair influence at a high probability. Further, we present an efficient algorithm that guarantees $1-1/e - \varepsilon$ approximation.

In this paper, we study the adversarial attacks on influence maximization under dynamic influence propagation models in social networks. In particular, given a known seed set S, the problem is to minimize the influence spread from S by deleting a limited number of nodes and edges. This problem reflects many application scenarios, such as blocking virus (e.g. COVID-19) propagation in social networks by quarantine and vaccination, blocking rumor spread by freezing fake accounts, or attacking competitor's influence by incentivizing some users to ignore the information from the competitor. In this paper, under the linear threshold model, we adapt the reverse influence sampling approach and provide efficient algorithms of sampling valid reverse reachable paths to solve the problem. We present three different design choices on reverse sampling, which all guarantee 1/2 - ε approximation (for any small ε >0) and an efficient running time.

In social network analysis, identifying the important nodes (key nodes) is a significant task in various applications. There are three most popular related tasks named influential node ranking, influence maximization, and network dismantling. Although these studies are different due to their own motivation, they share many similarities, which could confuse the non-domain readers and users. Moreover, few studies have explored the correlations between key nodes obtained from different tasks, hindering our further understanding of social networks. In this paper, we contribute to the field by conducting an in-depth survey of different kinds of key nodes through comparing these key nodes under our proposed framework and revealing their deep relationships. First, we clarify and formalize three existing popular studies under a uniform standard. Then we collect a group of crucial metrics and propose a fair comparison framework to analyze the features of key nodes identified by different research fields. From a large number of experiments and deep analysis on twenty real-world datasets, we not only explore correlations between key nodes derived from the three popular tasks, but also summarize insightful conclusions that explain how key nodes differ from each other and reveal their unique features for the corresponding tasks. Furthermore, we show that Shapley centrality could identify key nodes with more generality, and these nodes could also be applied to the three popular tasks simultaneously to a certain extent.

The influence are not linked to any footnote in the text. Please check and suggest. maximization problem aims at selecting the k most influential nodes (i.e., seed nodes) from a social network, where the nodes can maximize the number of influenced nodes activated by a certain propagation model. However, the widely used Independent Cascade model shares the same propagation probability among substantial adjacent node pairs, which is too idealistic and unreasonable in practice. In addition, most heuristic algorithms for influence maximization need to update the expected influence of the remaining nodes in the seed selection process, resulting in high computation cost. To address these non-trivial problems, we propose a novel edge propagation probability calculation method. The method first utilizes the triadic closure structure of social networks to precisely measure the closeness between nodes and assigns different propagation probabilities to each edge, deriving a Triadic Closure-based Independent Cascade (TC-IC) model. Then, we further propose a heuristic influence maximization algorithm named Triadic Closure-based Influence Maximization (TC-IM). The algorithm evaluates the expected influence of a node by integrating the triadic closure weighted propagation probability and the triadic closure weighted degree. Especially, in the seed selection process, only the most influential node that has not been updated in the current round needs to be updated, which significantly improves the efficiency. Besides, we further provide theoretical proofs to guarantee the correctness of this updating strategy. Experimental results on nine real datasets and three propagation models demonstrate that: (1) The TC-IC model can set a proper propagation probability for each node pair, where the IM algorithms could easily identify influential nodes; (2) The TC-IM algorithm can significantly reduce the complexity through an efficient updating strategy with a comparable influence spread to the approximation IM algorithms; (3) Besides, the TC-IM algorithm also exhibits stable performance under other IC models including UIC and WIC, exhibiting good stability and generality.