Behavior patterns of online users and the effect on information filtering

Understanding the structure and evolution of web-based user-item bipartite networks is an important task since they play a fundamental role in online information filtering. In this paper, we focus on investigating the patterns of online users’ behavior and the effect on recommendation process. Empirical analysis on the e-commercial systems show that users’ taste preferences are heterogeneous in general but their interests for niche items are highly clustered. Additionally, recommendation processes are investigated on both the real networks and the reshuffled networks in which real users’ behavior patterns can be gradually destroyed. We find that the performance of personalized recommendation methods is strongly related to the real network structure. Detailed study on each item shows that most hot items are accurately recommended and their recommendation accuracy is robust to the reshuffling process. However, the accuracy for niche items is relatively low and drops significantly after removing users’ behavior patterns. Our work is also meaningful in practical sense since it reveals an effective direction to improve the accuracy and the robustness of the existing recommender systems.     

Personalized recommendation via an improved NBI algorithm and user influence model in a Microblog network

Bipartite network based recommendations have attracted extensive attentions in recent years. Differing from traditional object-oriented recommendations, the recommendation in a Microblog network has two crucial differences. One is high authority users or one’s special friends usually play a very active role in tweet-oriented recommendation. The other is that the object in a Microblog network corresponds to a set of tweets on same topic instead of an actual and single entity, e.g. goods or movies in traditional networks. Thus repeat recommendations of the tweets in one’s collected topics are indispensable. Therefore, this paper improves network based inference (NBI) algorithm by original link matrix and link weight on resource allocation processes. This paper finally proposes the Microblog recommendation model based on the factors of improved network based inference and user influence model. Adjusting the weights of these two factors could generate the best recommendation results in algorithm accuracy and recommendation personalization.     

The scaling of human mobility by taxis is exponential

As a significant factor in urban planning, traffic forecasting and prediction of epidemics, modeling patterns of human mobility draws intensive attention from researchers for decades. Power-law distribution and its variations are observed from quite a few real-world human mobility datasets such as the movements of banking notes, trackings of cell phone users’ locations and trajectories of vehicles. In this paper, we build models for 20 million trajectories with fine granularity collected from more than 10 thousand taxis in Beijing. In contrast to most models observed in human mobility data, the taxis’ traveling displacements in urban areas tend to follow an exponential distribution instead of a power-law. Similarly, the elapsed time can also be well approximated by an exponential distribution. Worth mentioning, analysis of the interevent time indicates the bursty nature of human mobility, similar to many other human activities.     

A weighted network evolution with traffic flow

In this paper, we propose a simple weighted network model that generalizes the complex network model evolution with traffic flow previously presented to investigate the relationship between traffic flow and network structure. In the model, the nodes in the network are represented by the traffic flow states, the links in the network are represented by the transform of the traffic flow states, and the traffic flow transported when performing the transform of the traffic flow states is considered as the weight of the link. Several topological features of this generalized weighted model, such as the degree distribution and strength distribution, have been numerically studied. A scaling behavior between the strength and degree s∼klogk is obtained. By introducing some constraints to the generalized weighted model, we study its subnetworks and find that the scaling behavior between the strength and degree is conserved, though the topology properties are quite sensitive to the constraints.     

An empirical study of Chinese language networks

Chinese is spoken by the largest number of people in the world, and it is regarded as one of the most important languages. In this paper, we explore the statistical properties of Chinese language networks (CLNs) within the framework of complex network theory. Based on one of the largest Chinese corpora, i.e. People’s Daily Corpus, we construct two networks (CLN1 and CLN2) from two different respects, with Chinese words as nodes. In CLN1, a link between two nodes exists if they appear next to each other in at least one sentence; in CLN2, a link represents that two nodes appear simultaneously in a sentence. We show that both networks exhibit small-world effect, scale-free structure, hierarchical organization and disassortative mixing. These results indicate that in many topological aspects Chinese language shapes complex networks with organizing principles similar to other previously studied language systems, which shows that different languages may have some common characteristics in their evolution processes. We believe that our research may shed some new light into the Chinese language and find some potentially significant implications.     

Evolving model of weighted networks inspired by scientific collaboration networks

Inspired by scientific collaboration networks (SCN), especially our empirical analysis of econophysicists network, an evolutionary model for weighted networks is proposed. Besides a new vertex added in at every time step, old vertices can also attempt to build up new links, or to reconnect the existing links. The number of connections repeated between two nodes is converted into the weight of the link. This provides a natural way for the evolution of link weight. The path-dependent preferential attachment mechanism with local information is also introduced. It increases the clustering coefficient of the network significantly. The model shows the scale-free phenomena in degree and vertex weight distribution. It also gives well qualitatively consistent behavior with the empirical results.     

Inferring Users’ Social Roles with a Multi-Level Graph Neural Network Model

Users of social networks have a variety of social statuses and roles. For example, the users of Weibo include celebrities, government officials, and social organizations. At the same time, these users may be senior managers, middle managers, or workers in companies. Previous studies on this topic have mainly focused on using the categorical, textual and topological data of a social network to predict users’ social statuses and roles. However, this cannot fully reflect the overall characteristics of users’ social statuses and roles in a social network. In this paper, we consider what social network structures reflect users’ social statuses and roles since social networks are designed to connect people. Taking an Enron email dataset as an example, we analyzed a preprocessing mechanism used for social network datasets that can extract users’ dynamic behavior features. We further designed a novel social network representation learning algorithm in order to infer users’ social statuses and roles in social networks through the use of an attention and gate mechanism on users’ neighbors. The extensive experimental results gained from four publicly available datasets indicate that our solution achieves an average accuracy improvement of 2% compared with GraphSAGE-Mean, which is the best applicable inductive representation learning method.     

Interpersonal interactions and human dynamics in a large social network

We study a large social network consisting of over 10⁶ individuals, who form an Internet community and organize themselves in groups of different sizes. On the basis of the users’ list of friends and other data registered in the database we investigate the structure and time development of the network. The structure of this friendship network is very similar to the structure of different social networks. However, here a degree distribution exhibiting two scaling regimes, power-law for low connectivity and exponential for large connectivity, was found. The groups size distribution and distribution of number of groups of an individual have power-law form. We found very interesting scaling laws concerning human dynamics. Our research has shown how long people are interested in a single task.     

Revealing how network structure affects accuracy of link prediction

Link prediction plays an important role in network reconstruction and network evolution. The network structure affects the accuracy of link prediction, which is an interesting problem. In this paper we use common neighbors and the Gini coefficient to reveal the relation between them, which can provide a good reference for the choice of a suitable link prediction algorithm according to the network structure. Moreover, the statistical analysis reveals correlation between the common neighbors index, Gini coefficient index and other indices to describe the network structure, such as Laplacian eigenvalues, clustering coefficient, degree heterogeneity, and assortativity of network. Furthermore, a new method to predict missing links is proposed. The experimental results show that the proposed algorithm yields better prediction accuracy and robustness to the network structure than existing currently used methods for a variety of real-world networks.     

A network analysis of the Chinese stock market

In many practical important cases, a massive dataset can be represented as a very large network with certain attributes associated with its vertices and edges. Stock markets generate huge amounts of data, which can be use for constructing the network reflecting the market’s behavior. In this paper, we use a threshold method to construct China’s stock correlation network and then study the network’s structural properties and topological stability. We conduct a statistical analysis of this network and show that it follows a power-law model. We also detect components, cliques and independent sets in this network. These analyses allows one to apply a new data mining technique of classifying financial instruments based on stock price data, which provides a deeper insight into the internal structure of the stock market. Moreover, we test the topological stability of this network and find that it displays a topological robustness against random vertex failures, but it is also fragile to intentional attacks. Such a network stability property would be also useful for portfolio investment and risk management.     

An optimal network for passenger traffic

The optimal solution of an inter-city passenger transport network has been studied using Zipf’s law for the city populations and the Gravity law describing the fluxes of inter-city passenger traffic. Assuming a fixed value for the cost of transport per person per kilometer we observe that while the total traffic cost decreases, the total wiring cost increases with the density of links. As a result the total cost to maintain the traffic distribution is optimal at a certain link density which vanishes on increasing the network size. At a finite link density the network is scale-free. Using this model the air-route network of India has been generated and an one-to-one comparison of the nodal degree values with the real network has been made.     

Exploring the mobility of mobile phone users

Mobile phone datasets allow for the analysis of human behavior on an unprecedented scale. The social network, temporal dynamics and mobile behavior of mobile phone users have often been analyzed independently from each other using mobile phone datasets. In this article, we explore the connections between various features of human behavior extracted from a large mobile phone dataset. Our observations are based on the analysis of communication data of 100,000 anonymized and randomly chosen individuals in a dataset of communications in Portugal. We show that clustering and principal component analysis allow for a significant dimension reduction with limited loss of information. The most important features are related to geographical location. In particular, we observe that most people spend most of their time at only a few locations. With the help of clustering methods, we then robustly identify home and office locations and compare the results with official census data. Finally, we analyze the geographic spread of users’ frequent locations and show that commuting distances can be reasonably well explained by a gravity model.     

Optimal dynamic bandwidth allocation for complex networks

Traffic capacity of one network strongly depends on the link’s bandwidth allocation strategy. In previous bandwidth allocation mechanisms, once one link’s bandwidth is allocated, it will be fixed throughout the overall traffic transmission process. However, the traffic load of every link changes from time to time. In this paper, with finite total bandwidth resource of the network, we propose to dynamically allocate the total bandwidth resource in which each link’s bandwidth is proportional to the queue length of the output buffer of the link per time step. With plenty of data packets in the network, the traffic handling ability of all links of the network achieves full utilization. The theoretical analysis and the extensive simulation results on complex networks are consistent. This work is valuable for network service providers to improve network performance or to do reasonable network design efficiently.     

E-Learning Performance Prediction: Mining the Feature Space of Effective Learning Behavior

Learning analysis provides a new opportunity for the development of online education, and has received extensive attention from scholars at home and abroad. How to use data and models to predict learners’ academic success or failure and give teaching feedback in a timely manner is a core problem in the field of learning analytics. At present, many scholars use key learning behaviors to improve the prediction effect by exploring the implicit relationship between learning behavior data and grades. At the same time, it is very important to explore the association between categories and prediction effects in learning behavior classification. This paper proposes a self-adaptive feature fusion strategy based on learning behavior classification, aiming to mine the effective E-learning behavior feature space and further improve the performance of the learning performance prediction model. First, a behavior classification model (E-learning Behavior Classification Model, EBC Model) based on interaction objects and learning process is constructed; second, the feature space is preliminarily reduced by entropy weight method and variance filtering method; finally, combined with EBC Model and a self-adaptive feature fusion strategy to build a learning performance predictor. The experiment uses the British Open University Learning Analysis Dataset (OULAD). Through the experimental analysis, an effective feature space is obtained, that is, the basic interactive behavior (BI) and knowledge interaction behavior (KI) of learning behavior category has the strongest correlation with learning performance.And it is proved that the self-adaptive feature fusion strategy proposed in this paper can effectively improve the performance of the learning performance predictor, and the performance index of accuracy(ACC), F1-score(F1) and kappa(K) reach 98.44%, 0.9893, 0.9600. This study constructs E-learning performance predictors and mines the effective feature space from a new perspective, and provides some auxiliary references for online learners and managers.     

Analysis of price behavior in lazy $-game

A non-cooperative iterated multiagent game, called a minority game, and its variations have been extensively studied in this decade. To increase its market similarity, a $-game was presented by observing the current and the next agent’s payoffs. However, since the $-game is defined as an offline game, it is difficult to simulate it in practice. So we propose a new online version of the $-game, called a lazy $-game, and analyze the price behavior of the game. First, we reveal the condition of a bubble phenomenon in the lazy $-game. Next, we investigate the price behavior in the lazy $-game and show that there are some upper/lower bounds of the price as long as both the buyers group and the sellers group are nonempty. Then, we consider the similarity between the lazy $-game and the $-game. Finally, we present some simulation results.     

Emergence of scaling and assortative mixing through altruism

Many social networks, apart from displaying scale-free characteristics observed in some instances, possess another remarkable feature that distinguishes them from those that appear in biological and technological context—assortativity. However, little or no attention has been payed to the mechanism of assortativity in modeling these networks. Inspired by individuals’ altruistic behavior in sociology, we develop a model with a different growth mechanism called “altruistic attachment”, which can reproduce not only the power law degree distribution but degree correlations. We study in detail the statistical properties of our network model, which we also demonstrate striking differences with the BA model, and can portray real social networks more precisely.     

TLP-CCC: Temporal Link Prediction Based on Collective Community and Centrality Feature Fusion

In the domain of network science, the future link between nodes is a significant problem in social network analysis. Recently, temporal network link prediction has attracted many researchers due to its valuable real-world applications. However, the methods based on network structure similarity are generally limited to static networks, and the methods based on deep neural networks often have high computational costs. This paper fully mines the network structure information and time-domain attenuation information, and proposes a novel temporal link prediction method. Firstly, the network collective influence (CI) method is used to calculate the weights of nodes and edges. Then, the graph is divided into several community subgraphs by removing the weak link. Moreover, the biased random walk method is proposed, and the embedded representation vector is obtained by the modified Skip-gram model. Finally, this paper proposes a novel temporal link prediction method named TLP-CCC, which integrates collective influence, the community walk features, and the centrality features. Experimental results on nine real dynamic network data sets show that the proposed method performs better for area under curve (AUC) evaluation compared with the classical link prediction methods.     

How people make friends in social networking sites—A microscopic perspective

We study the detailed growth of a social networking site with full temporal information by examining the creation process of each friendship relation that can collectively lead to the macroscopic properties of the network. We first study the reciprocal behavior of users, and find that link requests are quickly responded to and that the distribution of reciprocation intervals decays in an exponential form. The degrees of inviters/accepters are slightly negatively correlative with reciprocation time. In addition, the temporal feature of the online community shows that the distributions of intervals of user behaviors, such as sending or accepting link requests, follow a power law with a universal exponent, and peaks emerge for intervals of an integral day. We finally study the preferential selection and linking phenomena of the social networking site and find that, for the former, a linear preference holds for preferential sending and reception, and for the latter, a linear preference also holds for preferential acceptance, creation, and attachment. Based on the linearly preferential linking, we put forward an analyzable network model which can reproduce the degree distribution of the network. The research framework presented in the paper could provide a potential insight into how the micro-motives of users lead to the global structure of online social networks.     

Other-regarding preference and the evolutionary prisoner’s dilemma on complex networks

Prevailing models of the evolutionary prisoner’s game on networks always assume that agents are pursuing their own profit maximization. But the results from experimental games show that many agents have other-regarding preference. In this paper, we study the emergence of cooperation from the prisoner’s dilemma game on complex networks while some agents exhibit other-regarding preference such as inequality aversion, envious and guilty emotions. Contrary to common ideas, the simulation results show that the existence of inequality aversion agents does not promote cooperation emergence on a BA (Barabási and Albert) scale-free network in most situations. If the defection attraction is big and agents exhibit strong preference for inequality aversion, the frequency of cooperators will be lower than in situations where no inequality aversion agents exist. In some cases, the existence of the inequality agents will even induce the frequency of cooperators to zero, a feature which is not observed in previous research on the prisoner’s dilemma game when the underlying interaction topology is a BA scale-free network. This means that if an agent cares about equality too much, it will be difficult for cooperation to emerge and the frequency of cooperators will be low on BA networks. The research on the effect of envy or guilty emotions on the emergence of cooperation in the prisoner’s dilemma game on BA networks obtains similar results, though some differences exist. However, simulation results on a WS (Watts and Strogatz) small-world network display another scenario. If agents care about the inequality of agents very much, the WS network favors cooperation emergence in the prisoners’ dilemma game when other-regarding agents exist. If the agent weight on other-regarding is lowered, the cooperation frequencies emerging on a WS network are not much different from those in situations without other-regarding agents, although the frequency of cooperators is lower than those of the situation without other-regarding preference agents sometimes. All the simulation results imply that inequality aversion and its variations can have important effects on cooperation emergence in the prisoner’s dilemma game, and different network topologies have different effects on cooperation emergence in the prisoner’s dilemma game played on complex networks.