In: PLoS ONE, 2013, vol. 8, no. 8, p. e70094
Accuracy and diversity are two important aspects to evaluate the performance of recommender systems. Two diffusion-based methods were proposed respectively inspired by the mass diffusion (MD) and heat conduction (HC) processes on networks. It has been pointed out that MD has high recommendation accuracy yet low diversity, while HC succeeds in seeking out novel or niche items but with relatively...
|
In: Physical Review E, 2013, vol. 88, no. 1, p. 012818
Some epidemic spreading models are usually applied to analyze the propagation of opinions or news. However, the dynamics of epidemic spreading and information or behavior spreading are essentially different in many aspects. Centola's experiments [ Science 329 1194 (2010)] on behavior spreading in online social networks showed that the spreading is faster and broader in regular networks than in...
|
In: PLoS ONE, 2013, vol. 8, no. 2, p. e55437
Uncovering factors underlying the network formation is a long-standing challenge for data mining and network analysis. In particular, the microscopic organizing principles of directed networks are less understood than those of undirected networks. This article proposes a hypothesis named potential theory, which assumes that every directed link corresponds to a decrease of a unit potential and...
|
In: Physica A: Statistical Mechanics and its Applications, 2012, vol. 139, no. 22, p. 5769–5778
To evaluate the performance of prediction of missing links, the known data are randomly divided into two parts, the training set and the probe set. We argue that this straightforward and standard method may lead to terrible bias, since in real biological and information networks, missing links are more likely to be links connecting low-degree nodes. We therefore study how to uncover missing links...
|
In: Physics Reports, 2012, vol. 519, no. 1, p. 1–49
The ongoing rapid expansion of the Internet greatly increases the necessity of effective recommender systems for filtering the abundant information. Extensive research for recommender systems is conducted by a broad range of communities including social and computer scientists, physicists, and interdisciplinary researchers. Despite substantial theoretical and practical achievements, unification...
|
In: New Journal of Physics, 2012, vol. 14, p. 033033
The number of citations is a widely used metric for evaluating the scientific credit of papers, scientists and journals. However, it so happens that papers with fewer citations from prestigious scientists have a higher influence than papers with more citations. In this paper, we argue that by whom the paper is being cited is of greater significance than merely the number of citations....
|
In: Physical Review E - Statistical, Nonlinear, and Soft Matter Physics, 2012, vol. 85, no. 4, p. 046108
The advent of the Internet and World Wide Web has led to unprecedent growth of the information available. People usually face the information overload by following a limited number of sources which best fit their interests. It has thus become important to address issues like who gets followed and how to allow people to discover new and better information sources. In this paper we conduct an...
|
In: Physica A: Statistical Mechanics and its Applications, 2012, vol. 391, no. 3, p. 647–655
Recently, supply networks have attracted increasing attention from the scientific community. However, it lacks serious consideration of social preference in Supply Chain Management. In this paper, we develop an evolutionary decision-making model to characterize the effects of suppliers’ altruism in supply networks, and find that the performances of both suppliers and supply chains are improved...
|
In: New Journal of Physics, 2011, vol. 13, p. 123005
The spreading dynamics of information and diseases are usually analyzed by using a unified framework and analogous models. In this paper, we propose a model to emphasize the essential difference between information spreading and epidemic spreading, where the memory effects, the social reinforcement and the non-redundancy of contacts are taken into account. Under certain conditions, the...
|
In: Europhysics Letters, 2011, vol. 96, no. 4, p. 48007
The common-neighbor–based method is simple yet effective to predict missing links, which assume that two nodes are more likely to be connected if they have more common neighbors. In the traditional method, each common neighbor of two nodes contributes equally to the connection likelihood. In this letter, we argue that different common neighbors may play different roles and thus contributes...
|
