Federated Learning (FL) enables collaborative model training across distributed clients without requiring the exchange of raw data. However, existing One-Shot FL (OSFL) methods, designed for communication efficiency by reducing fed-erated rounds to one, suffer substantial performance degradation when faced with highly non-IID data across clients, primarily due to critical distribution shifts: label shift, feature shift, and concept shift. In this paper, we introduce POSFed, a new personalized three-stage approach, to systematically address these fundamen-tal limitations: (1) Each client locally generates robust and label-agnostic synthetic datasets via self-supervising learning, ensuring essential knowledge is captured despite local distribution shifts; (2) The server aggregates all synthetic datasets to train a global feature extractor, capturing generalizable and transferable rep-resentations across heterogeneous client data; and (3) Each client efficiently adapts the feature extractor by learning a personalized classification head on its own data, enabling effective local customization and mitigating both feature and concept shifts. Extensive experiments across multiple benchmarks demonstrate that POSFed significantly outperforms state-of-the-art methods, achieving performance comparable to multi-round personalized approaches while using only one communication round. By ensuring both superior personalization and practical communication efficiency, POSFed establishes a feasible paradigm for FL under more realistic and heterogeneous conditions. Code is available at https://github.com/I643204431IPOSFed.

Influence Maximization (IM) is a classical combinatorial optimization problem, which can be widely used in mobile networks, social computing, and recommendation systems. It aims at selecting a small number of users such that maximizing the influence spread across the online social network. Because of its potential commercial and academic value, there are a lot of researchers focusing on studying the IM problem from different perspectives. The main challenge comes from the NP-hardness of the IM problem and #P-hardness of estimating the influence spread, thus traditional algorithms for overcoming them can be categorized into two classes: heuristic algorithms and approximation algorithms. However, there is no theoretical guarantee for heuristic algorithms, and the theoretical design is close to the limit. Therefore, it is almost impossible to further optimize and improve their performance. With the rapid development of artificial intelligence, technologies based on Machine Learning (ML) have achieved remarkable achievements in many fields. In view of this, in recent years, a number of new methods have emerged to solve combinatorial optimization problems by using ML-based techniques. These methods have the advantages of fast solving speed and strong generalization ability to unknown graphs, which provide a brand-new direction for solving combinatorial optimization problems. Therefore, we abandon the traditional algorithms based on iterative search and review the recent development of ML-based methods, especially Deep Reinforcement Learning, to solve the IM problem and other variants in social networks. We focus on summarizing the relevant background knowledge, basic principles, common methods, and applied research. Finally, the challenges that need to be solved urgently in future IM research are pointed out.

Influence maximization problem attempts to find a small subset of nodes that makes the expected influence spread maximized, which has been researched intensively before. They all assumed that each user in the seed set we select is activated successfully and then spread the influence. However, in the real scenario, not all users in the seed set are willing to be an influencer. Based on that, we consider each user associated with a probability with which we can activate her as a seed, and we can attempt to activate her many times. In this article, we study the adaptive influence maximization with multiple activations (Adaptive-IMMA) problem, where we select a node in each iteration, observe whether she accepts to be a seed, if yes, wait to observe the influence diffusion process; if no, we can attempt to activate her again with a higher cost or select another node as a seed. We model the multiple activations mathematically and define it on the domain of integer lattice. We propose a new concept, adaptive dr-submodularity, and show our Adaptive-IMMA is the problem that maximizing an adaptive monotone and dr-submodular function under the expected knapsack constraint. Adaptive dr-submodular maximization problem is never covered by any existing studies. Thus, we summarize its properties and study its approximability comprehensively, which is a non-trivial generalization of existing analysis about adaptive submodularity. Besides, to overcome the difficulty to estimate the expected influence spread, we combine our adaptive greedy policy with sampling techniques without losing the approximation ratio but reducing the time complexity. Finally, we conduct experiments on several real datasets to evaluate the effectiveness and efficiency of our proposed policies.

Cyberbullying has caused serious consequences especially for social network users in recent years. However, the challenge is how to fight against the cyberbullying effectively from the algorithmic perspective. In this article, we study the fighting against the cyberbullying problem, i.e., identify an initial witness set with a budget to spread the positive influence to protect the users in a specific target set such that the number of cybervictim users in the target set being activated by the seed set of cyberbullying is minimized. We first formulate this problem and show its NP-hardness. We further prove that the objective function is submodular with respect to the size of witnesses set when we convert the original problem into the maximal version. Then we propose a stochastic approach to solve this maximal version problem based on the Reverse Sampling Technique with a constant factor guarantee. In addition, we provide theoretical analysis and discuss the relationship between the optimal value and the value returned by the proposed algorithm. To evaluate the proposed approach, we implement extensive experiments on synthetic and real datasets. The experimental results show our approach is superior to the comparison methods.

Influence maximization problem attempts to find a small subset of nodes in a social network that makes the expected influence maximized, which has been researched intensively before. Most of the existing literature focus only on maximizing total influence, but it ignores whether the influential distribution is balanced through the network. Even though the total influence is maximized, but gathered in a certain area of social network. Sometimes, this is not advisable. In this article, we propose a novel seeding strategy based on community structure, and formulate the Influence Maximization with Community Budget (IMCB) problem. In this problem, the number of seed nodes in each community is under the cardinality constraint, which can be classified as the problem of monotone submodular maximization under the matroid constraint. To give a satisfactory solution for IMCB problem under the triggering model, we propose the IMCB-Framework, which is inspired by the idea of continuous greedy process and pipage rounding, and derive the best approximation ratio for this problem. In IMCB-Framework, we adopt sampling techniques to overcome the high complexity of continuous greedy. Then, we propose a simplified pipage rounding algorithm, which reduces the complexity of IMCB-Framework further. Finally, we conduct experiments on three real-world datasets to evaluate the correctness and effectiveness of our proposed algorithms, as well as the advantage of IMCB-Framework against classical greedy method.

In recent years, social networks have become important platforms for people to disseminate information. However, we need to take effective measures such as blocking a set of links to control the negative rumors spreading over the network. In this article, we propose a Rumor Spread Minimization (RSM) problem, i.e., we remove an edge set from network such that the rumor spread is minimized. We first prove the objective function of RSM problem is not submodular. Then, we propose both submodular lower-bound and upper-bound of the objective function. Next, we develop a heuristic algorithm to approximate the objective function. Furthermore, we reformulate our objective function as the DS function (the Difference of Submodular functions). Finally, we conduct experiments on real-world datasets to evaluate our proposed method. The experiment results show that the upper and lower bounds are very close, which indicates the good quality of them. And, the proposed method outperforms the comparison methods.

We study the problem of active friending in online social networks. Given an initiator who want to friend a target person on a social network, we propose a strategy to support active friending through a series of recommendation lists. The lists serve as a step-to-step guidance for the initiator. We formulate an optimization problem, Constrained Active Friending CAF), for configuring the recommendation lists in the active friending process. Our goal is to maximize the acceptance probability of the invitation from the initiator to the friending target, by recommending selective intermediate friends to approach the target. We prove that CAF problem is NP-hard under the linear threshold model. We propose an algorithm based on discrete super-differentials that derives a guaranteed approximation for this problem. Extensive evaluation results on benchmark social network datasets validate the effectiveness and efficiency of our algorithms.

Social influence has been widely studied in areas of viral marketing, information diffusion and health care. Currently, most influence models only deal with a single influence without the interference of other influences. Also, the influence spreading in previous models must be triggered by individuals who have been activated by the influence. In this paper, we argue that it is the attraction from a specific influence makes an individual choose to spread it among multiple influences. Inspired by charged system theory in physics, a new influence model is proposed, considering individual features and social structure features. It also gives a natural description about how individuals make decisions among multiple influences. Then a novel algorithm based on this model is provided to predict human behavior. Extensive experiments on three real-world datasets demonstrate that our model and algorithm statistically outperform the state-of-the-art methods in terms of prediction accuracy.

Influence maximization problem is to find a set of seeds in social networks such that the cascade influence is maximized. Traditional models assume all nodes are willing to spread the influence once they are influenced, and they ignore the disparity between influence and profit of a product. In this paper by considering the role that price plays in viral marketing, we propose price related (PR) frame that contains PR-I and PR-L models for classic IC and LT models respectively, which is a pioneer work. We find that influence and profit are like two sides of the coin, high price hinders the influence propagation and to enlarge the influence some sacrifice on profit is inevitable. We propose Balanced Influence and Profit (BIP) maximization problem. We prove the NP-hardness of BIP maximization under PR-I and PR-L model. Unlike influence maximization, the BIP objective function is not monotone. Despite the non-monotony, we show BIP objective function is sub modular under certain conditions. Two unbudgeted greedy algorithms separately are devised. We conduct simulations on real-world datasets and evaluate the superiority of our algorithms over existing ones.

In this study, we formalize a multi-focal learning problem, where training data are partitioned into several different focal groups and the prediction model will be learned within each focal group. The multi-focal learning problem is motivated by numerous real-world learning applications. For instance, for the same type of problems encountered in a customer service center, the problem descriptions from different customers can be quite different. The experienced customers usually give more precise and focused descriptions about the problem. In contrast, the inexperienced customers usually provide more diverse descriptions. In this case, the examples from the same class in the training data can be naturally in different focal groups. As a result, it is necessary to identify those natural focal groups and exploit them for learning at different focuses. The key developmental challenge is how to identify those focal groups in the training data. As a case study, we exploit multi-focal learning for profiling problems in customer service centers. The results show that multifocal learning can significantly boost the learning accuracies of existing learning algorithms, such as Support Vector Machines (SVMs), for classifying customer problems.

This paper targets the identification of outlying sensors (that is, outlying reading sensors) and the detection of the reach of events in sensor networks. Typical applications include the detection of the transportation front line of some vegetation or animalcule's growth over a certain geographical region. We propose and analyze two novel algorithms for outlying sensor identification and event boundary detection. These algorithms are purely localized and, thus, scale well to large sensor networks. Their computational overhead is low, since only simple numerical operations are involved. Simulation results indicate that these algorithms can clearly detect the event boundary and can identify outlying sensors with a high accuracy and a low false alarm rate when as many as 20 percent sensors report outlying readings. Our work is exploratory in that the proposed algorithms can accept any kind of scalar values as inputs-a dramatic improvement over existing work, which takes only 0/1 decision predicates. Therefore, our algorithms are generic. They can be applied as long as "events" can be modeled by numerical numbers. Though designed for sensor networks, our algorithms can be applied to the outlier detection and regional data analysis in spatial data mining.

Data compression is an effective technique to improve the performance of data warehouses. Since cube operation represents the core of online analytical processing in data warehouses, it is a major challenge to develop efficient algorithms for computing cube on compressed data warehouses. To our knowledge, very few cube computation techniques have been proposed for compressed data warehouses to date in the literature. This paper presents a novel algorithm to compute cubes on compressed data warehouses. The algorithm operates directly on compressed data sets without the need of first decompressing them. The algorithm is applicable to a large class of mapping complete data compression methods. The complexity of the algorithm is analyzed in detail. The analytical and experimental results show that the algorithm is more efficient than all other existing cube algorithms. In addition, a heuristic algorithm to generate an optimal plan for computing cube is also proposed.

1 Introduction Widespread use of spatial databases[24] is leading to an increasing interest in mining interesting and useful but implicit spatial patterns[14, 17, 10, 22]. Efficient tools for extracting information from geo-spatial data, the focus of this work, are crucial to organizations which make decisions based on large spatial data sets. These organizations are spread across many domains including ecology and environment management, public safety, transportation, public health, business, travel and tourism[2, 12].