Modeling complex networks with accelerating growth and aging effect

Liu, Jin; Li, Jian; Chen, Yadang; Chen, Xianyi; Zhou, Zhili; Yang, Zejun; Zhang, Chengjun

doi:10.1016/j.physleta.2019.02.004

Cited by 13 publications

(8 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, recommendation algorithms based on complex networks [24], especially bipartite graphs, have attracted increasing attention from scholars. This type of algorithm draws on the ideas of mass diffusion and heat conduction to abstract the input data of the recommendation system into complex network models.…”

Section: Related Workmentioning

confidence: 99%

Fusion of Internal Similarity to Improve the Accuracy of Recommendation Algorithm

Dhanabal¹,

Baskar²,

Premkumar³

2022

JUSST

View full text Add to dashboard Cite

Collaborative filtering algorithms (CF) and mass diffusion (MD) algorithms have been successfully applied to recommender systems for years and can solve the problem of information overload. However, both algorithms suffer from data sparsity, and both tend to recommend popular products, which have poor diversity and are not suitable for real life. In this paper, we propose a user internal similarity-based recommendation algorithm (UISRC). UISRC first calculates the item-item similarity matrix and calculates the average similarity between items purchased by each user as the user’s internal similarity. The internal similarity of users is combined to modify the recommendation score to make score predictions and suggestions. Simulation experiments on RYM and Last.FM datasets, the results show that UISRC can obtain better recommendation accuracy and a variety of recommendations than traditional CF and MD algorithms.

show abstract

Section: Related Workmentioning

confidence: 99%

Fusion of Internal Similarity to Improve the Accuracy of Recommendation Algorithm

Dhanabal¹,

Baskar²,

Premkumar³

2022

JUSST

View full text Add to dashboard Cite

show abstract

“…At the same time, the scale-free network models were developed, which leverage the mechanism of preferential connection, redirection or copy to interpret the power-law property of the node degree distribution. In recent years, following the above typical models, new mechanisms represented by weight [6][7][8][9], local world [10,11], nonlinear growth [11][12][13][14], location information [15][16][17][18], popularity and homophily [19][20][21], and triangle closure [22][23][24][25][26] are used to construct network evolution models.…”

Section: Introductionmentioning

confidence: 99%

Modeling Complex Networks Based on Deep Reinforcement Learning

et al. 2022

View full text Add to dashboard Cite

The network topology of complex networks evolves dynamically with time. How to model the internal mechanism driving the dynamic change of network structure is the key problem in the field of complex networks. The models represented by WS, NW, BA usually assume that the evolution of network structure is driven by nodes’ passive behaviors based on some restrictive rules. However, in fact, network nodes are intelligent individuals, which actively update their relations based on experience and environment. To overcome this limitation, we attempt to construct a network model based on deep reinforcement learning, named as NMDRL. In the new model, each node in complex networks is regarded as an intelligent agent, which reacts with the agents around it for refreshing its relationships at every moment. Extensive experiments show that our model not only can generate networks owing the properties of scale-free and small-world, but also reveal how community structures emerge and evolve. The proposed NMDRL model is helpful to study propagation, game, and cooperation behaviors in networks.

show abstract

“…While various rules have been proposed to explain the topology of real networks [10], most models assume a constant rate of network growth, i.e., the addition of a fixed number of nodes at each time step [15,20,21]. However, the results of empirical analysis of numerous technological and social systems show that their growth is time-dependent [23,24,25,26]. The accelerated growth in complex networks is the cause of the high heterogeneity in the distribution of webpages among websites [23] and the emergence of highly cited authors in citation networks [26].…”

Section: Introductionmentioning

confidence: 99%

Growth signals determine the topology of evolving networks

Vranić¹,

Dankulov²

2020

Preprint

View full text Add to dashboard Cite

Network science provides an indispensable theoretical framework for studying the structure and function of real complex systems. Different network models are often used for finding the rules that govern their evolution, whereby the correct choice of model details is crucial for obtaining relevant insights. We here study how the structure of networks generated with the aging nodes model depends on the properties of the growth signal. We use different fluctuating signals and compare structural dissimilarities of the networks with those obtained with a constant growth signal. We show that networks with power-law degree distributions, which are obtained with time-varying growth signals, are correlated and clustered, while networks obtained with a constant growth signal are not. Indeed, the properties of the growth signal significantly determine the topology of the obtained networks and thus ought to be considered prominently in models of complex systems.

show abstract

Modeling complex networks with accelerating growth and aging effect

Cited by 13 publications

References 23 publications

Fusion of Internal Similarity to Improve the Accuracy of Recommendation Algorithm

Fusion of Internal Similarity to Improve the Accuracy of Recommendation Algorithm

Modeling Complex Networks Based on Deep Reinforcement Learning

Growth signals determine the topology of evolving networks

Contact Info

Product

Resources

About