博文
写综述就是成全别人，恶心自己
||
前几天无意中打开Physica A首页，发现2011年写的那篇关于复杂网络链路预测的综述还在首页挂着，一年以前就在那了，如今下载量已荣升首位，引用在ScienceDirect数据库中也是第三名，着实感叹了一下。不是惊叹欢喜自己的工作受到广泛关注，而是一丝酸楚流入心头啊~
综述无疑是外部性最大的一种款式，但写的时候却极其痛苦，篇幅长不说，还要字斟句酌，对这个方向有整体的规划和把握，难难难，总之就是耗时费力往往还不算是科研成果。除了上面关于链路预测的综述，我们团队在2012年在Phys Rep发表的关于推荐系统的综述也颇受关注，如今引用已超过70次。于是在复杂网络信息挖掘方向上就差一篇ranking了，经过一年多的努力，这篇中文综述终于在前几日完稿！发出后还请大家批评指正
链路预测综述：http://www.sciencedirect.com/science/article/pii/S037843711000991X
推荐系统综述：http://www.sciencedirect.com/science/article/pii/S0370157312000828
相关博文：http://blog.sciencenet.cn/home.php?mod=space&uid=3075&do=blog&id=772032
15 March 2011
Linyuan Lü | Tao Zhou
Abstract: Link prediction in complex networks has attracted increasing attention from both physical and computer science communities. The algorithms can be used to extract missing information, identify spurious interactions, evaluate network evolving mechanisms, and so on. This article summaries recent progress about link prediction algorithms, emphasizing on the contributions from physical perspectives and approaches, such as the random-walk-based methods and the maximum likelihood methods. We also introduce three typical applications: reconstruction of networks, evaluation of network evolving mechanism and classification of partially labeled networks. Finally, we introduce some applications and outline future challenges of link prediction algorithms.
1 December 2006
Partha Chakroborty
Abstract: The aim of the paper is to present an engineer's viewpoint of traffic streams and their models both at the macroscopic and microscopic levels. The paper concentrates on two classes of macroscopic models (namely, stream description models and travel time estimation models). At the microscopic level the paper concentrates on car-following models and also presents a relatively recent idea on developing a comprehensive microscopic model of driver behaviour. Finally, the paper presents some properties which all microscopic models of traffic flow should possess and also tries to identify areas where research will bring about a qualitative jump in the understanding of how traffic flows.
1 October 2008
Alexandre Souto Martinez | Rodrigo Silva González | César Augusto Sangaletti Terçariol
Abstract: Consider the one-parameter generalizations of the logarithmic and exponential functions which are obtained from the integration of non-symmetrical hyperboles. These generalizations coincide to the one obtained in the context of non-extensive thermostatistics. We show that these functions are suitable to describe and unify the great majority of continuous growth models, which we briefly review. Physical interpretation to the generalization function parameter is given for the Richards’ model, which has an underlying microscopic model to justify it.
15 August 2002
A.L Barabási | H Jeong | Z Néda | E Ravasz | A Schubert | T Vicsek
Abstract: The co-authorship network of scientists represents a prototype of complex evolving networks. In addition, it offers one of the most extensive database to date on social networks. By mapping the electronic database containing all relevant journals in mathematics and neuro-science for an 8-year period (1991–98), we infer the dynamic and the structural mechanisms that govern the evolution and topology of this complex system. Three complementary approaches allow us to obtain a detailed characterization. First, empirical measurements allow us to uncover the topological measures that characterize the network at a given moment, as well as the time evolution of these quantities. The results indicate that the network is scale-free, and that the network evolution is governed by preferential attachment, affecting both internal and external links. However, in contrast with most model predictions the average degree increases in time, and the node separation decreases. Second, we propose a simple model that captures the network's time evolution. In some limits the model can be solved analytically, predicting a two-regime scaling in agreement with the measurements. Third, numerical simulations are used to uncover the behavior of quantities that could not be predicted analytically. The combined numerical and analytical results underline the important role internal links play in determining the observed scaling behavior and network topology. The results and methodologies developed in the context of the co-authorship network could be useful for a systematic study of other complex evolving networks as well, such as the world wide web, Internet, or other social networks.
1 February 2014
Javier E. Contreras-Reyes
Abstract: An asymptotic expression for the Kullback–Leibler (KL) divergence measure of multivariate skew-t distributions (MST) is derived. This novel class of flexible family distributions incorporates a shape and degree of freedom parameters, in order to manipulate the skewness and heavy-tail presence of the data, respectively. The quadratic form expressions of MST models are used to provide asymptotic measures. Additional inequalities for MST entropy and simulation studies are reported. Finally, the expected values of the KL divergence of a sample correlation matrix obtained by Pearson’s correlation coefficient are discussed.
15 February 2014
Giuliano Andrea Pagani | Marco Aiello
Abstract: The shift towards an energy grid dominated by prosumers (consumers and producers of energy) will inevitably have repercussions on the electricity distribution infrastructure. Today the grid is a hierarchical one delivering energy from large scale facilities to end-users. Tomorrow it will be a capillary infrastructure at the medium and low voltage levels that will support local energy trading among prosumers. We investigate how different network topologies and growth models facilitate a more efficient and reliable network, and how they can facilitate the emergence of a decentralized electricity market. We show how connectivity plays an important role in improving the properties of reliability and path-cost reduction. Our results indicate that a specific type of evolution balances best the ratio between increased connectivity and costs to achieve the network growth.
15 August 2013
Zhiguo Zhu
Abstract: The target of viral marketing on the platform of popular online social networks is to rapidly propagate marketing information at lower cost and increase sales, in which a key problem is how to precisely discover the most influential users in the process of information diffusion. A novel method is proposed in this paper for helping companies to identify such users as seeds to maximize information diffusion in the viral marketing. Firstly, the user trust network oriented to viral marketing and users’ combined interest degree in the network including isolated users are extensively defined. Next, we construct a model considering the time factor to simulate the process of information diffusion in viral marketing and propose a dynamic algorithm description. Finally, experiments are conducted with a real dataset extracted from the famous SNS website Epinions. The experimental results indicate that the proposed algorithm has better scalability and is less time-consuming. Compared with the classical model, the proposed algorithm achieved a better performance than does the classical method on the two aspects of network coverage rate and time-consumption in our four sub-datasets.
15 August 2012
Amanda L. Traud | Peter J. Mucha | Mason A. Porter
Abstract: We study the social structure of Facebook “friendship” networks at one hundred American colleges and universities at a single point in time, and we examine the roles of user attributes–gender, class year, major, high school, and residence–at these institutions. We investigate the influence of common attributes at the dyad level in terms of assortativity coefficients and regression models. We then examine larger-scale groupings by detecting communities algorithmically and comparing them to network partitions based on user characteristics. We thereby examine the relative importance of different characteristics at different institutions, finding for example that common high school is more important to the social organization of large institutions and that the importance of common major varies significantly between institutions. Our calculations illustrate how microscopic and macroscopic perspectives give complementary insights on the social organization at universities and suggest future studies to investigate such phenomena further.
1 August 2013
Sergio Rinaldi | Fabio Della Rossa | Pietro Landi
Abstract: We develop a mathematical model for mimicking the love story between Scarlett and Rhett described in “Gone with the Wind”. In line with tradition in classical physics, the model is composed of two Ordinary Differential Equations, one for Scarlett and one for Rhett, which encapsulate their main psycho-physical characteristics. The two lovers are described as so-called insecure individuals because they respond very strongly to small involvements of the partner but then attenuate their reaction when the pressure exerted by the partner becomes too high. These characteristics of Scarlett and Rhett clearly emerge during the first part of the film and are sufficient to develop a model that perfectly predicts the complex evolution and the dramatic end of the love story. Since the predicted evolution of the romantic relationship is a direct consequence of the characters of the two individuals, the agreement between the model and the film supports the high credibility of the story. Although credibility of a fictitious story is not necessary from a purely artistic point of view, in most cases it is very appreciated, at the point of being essential in making the film popular. In conclusion, we can say that we have explained with a scientific approach why “Gone with the Wind” has become one of the most successful films of all times.
1 April 2013
Qiang Yan | Lianren Wu | Lan Zheng
Abstract: The influence of microblog on information transmission is becoming more and more obvious. By characterizing the behavior of following and being followed as out-degree and in-degree respectively, a microblog social network was built in this paper. It was found to have short diameter of connected graph, short average path length and high average clustering coefficient. The distributions of out-degree, in-degree and total number of microblogs posted present power-law characters. The exponent of total number distribution of microblogs is negatively correlated with the degree of each user. With the increase of degree, the exponent decreases much slower. Based on empirical analysis, we proposed a social network based human dynamics model in this paper, and pointed out that inducing drive and spontaneous drive lead to the behavior of posting microblogs. The simulation results of our model match well with practical situation.
1 December 2013
Yutaka Tachimori | Hiroaki Iwanaga | Takashi Tahara
Abstract: Here, we constructed and analyzed a network (henceforth, “medical knowledge network”) derived from a commonly used medical text. We show that this medical knowledge network has small-world, scale-free, and hierarchical features. We then constructed a network from data from a hospital information system that reflected actual clinical practice and found that this network also had small-world, scale-free, and hierarchical features. Moreover, we found that both the diagnosis frequency distribution of the hospital network and the diagnosis degree distribution of the medical knowledge network obeyed a similar power law. These findings suggest that the structure of clinical practice may emerge from the mutual influence of medical knowledge and clinical practice, and that the analysis of a medical knowledge network may facilitate the investigation of the characteristics of medical practice.
1 February 2014
Wanqiu Guan | Haoyu Gao | Mingmin Yang | Yuan Li | Haixin Ma | Weining Qian | Zhigang Cao | Xiaoguang Yang
Abstract: The spread and resonance of users’ opinions on Sina Weibo, the most popular micro-blogging website in China, are tremendously influential, having significantly affected the processes of many real-world hot social events. We select 21 hot events that were widely discussed on Sina Weibo in 2011, and do some statistical analyses. Our main findings are that (i) male users are more likely to be involved, (ii) messages that contain pictures and those posted by verified users are more likely to be reposted, while those with URLs are less likely, (iii) the gender factor, for most events, presents no significant difference in reposting likelihood.
15 February 2012
Duanbing Chen | Linyuan Lü | Ming-Sheng Shang | Yi-Cheng Zhang | Tao Zhou
Abstract: Identifying influential nodes that lead to faster and wider spreading in complex networks is of theoretical and practical significance. The degree centrality method is very simple but of little relevance. Global metrics such as betweenness centrality and closeness centrality can better identify influential nodes, but are incapable to be applied in large-scale networks due to the computational complexity. In order to design an effective ranking method, we proposed a semi-local centrality measure as a tradeoff between the low-relevant degree centrality and other time-consuming measures. We use the Susceptible–Infected–Recovered (SIR) model to evaluate the performance by using the spreading rate and the number of infected nodes. Simulations on four real networks show that our method can well identify influential nodes.
1 October 1999
Albert-László Barabási | Réka Albert | Hawoong Jeong
Abstract: Random networks with complex topology are common in Nature, describing systems as diverse as the world wide web or social and business networks. Recently, it has been demonstrated that most large networks for which topological information is available display scale-free features. Here we study the scaling properties of the recently introduced scale-free model, that can account for the observed power-law distribution of the connectivities. We develop a mean-field method to predict the growth dynamics of the individual vertices, and use this to calculate analytically the connectivity distribution and the scaling exponents. The mean-field method can be used to address the properties of two variants of the scale-free model, that do not display power-law scaling.
1 March 2014
Lovro Šubelj | Marko Bajec
Abstract: Complex real-world networks commonly reveal characteristic groups of nodes like communities and modules. These are of value in various applications, especially in the case of large social and information networks. However, while numerous community detection techniques have been presented in the literature, approaches for other groups of nodes are relatively rare and often limited in some way. We present a simple propagation-based algorithm for general group detection that requires no a priori knowledge and has near ideal complexity. The main novelty here is that different types of groups are revealed through an adequate hierarchical group refinement procedure. The proposed algorithm is validated on various synthetic and real-world networks, and rigorously compared against twelve other state-of-the-art approaches on group detection, hierarchy discovery and link prediction tasks. The algorithm is comparable to the state of the art in community detection, while superior in general group detection and link prediction. Based on the comparison, we also discuss some prominent directions for future work on group detection in complex networks.
15 February 2010
Jianmei Yang | Canzhong Yao | Weicheng Ma | Guanrong Chen
Abstract: Buzzword-based viral marketing, known also as digital word-of-mouth marketing, is a marketing mode attached to some carriers on the Internet, which can rapidly copy marketing information at a low cost. Viral marketing actually uses a pre-existing social network where, however, the scale of the pre-existing network is believed to be so large and so random, so that its theoretical analysis is intractable and unmanageable. There are very few reports in the literature on how to design a spreading scheme for viral marketing on real social networks according to the traditional marketing theory or the relatively new network marketing theory. Complex network theory provides a new model for the study of large-scale complex systems, using the latest developments of graph theory and computing techniques. From this perspective, the present paper extends the complex network theory and modeling into the research of general viral marketing and develops a specific spreading scheme for viral marking and an approach to design the scheme based on a real complex network on the QQ instant messaging system. This approach is shown to be rather universal and can be further extended to the design of various spreading schemes for viral marketing based on different instant messaging systems.
15 October 2013
Shuichao Zhang | Gang Ren | Renfa Yang
Abstract: The mixed bicycle flow refers to the bicycle flow containing electric bicycles. The traffic characteristics data of the mixed bicycle flow was collected by the virtual coil method in Nanjing and Ningbo, China. And the speed–density characteristics of the mixed bicycle flow with different proportions of electric bicycles were obtained. The results show that the overall speed of the mixed bicycle flow containing electric bicycles is higher than that of pure bicycle flow when the density is relatively low. The speed decreases when the density is higher than 0.08 bic/m2; the speed–density characteristics of the bicycles and the electric bicycles tend to be the same when the density is higher than 0.25 bic/m2. And when the density reaches 0.58 bic/m2, the mixed bicycle flow becomes blocked and the speed is zero. The cellular automata model and gas dynamics model were also adopted to simulate the speed–density characteristics of the mixed bicycle flow. The simulation results of the cellular automata model are effectively consistent with the actual survey data when the density is lower than 0.225 bic/m2; the simulation results of the gas dynamics model are effectively consistent with the actual survey data when the density is higher than 0.300 bic/m2; but both of the two types of simulation models are inapplicable when the density is between 0.225 and 0.300 bic/m2. These results will be used in the management of mixed bicycles and the research of vehicle–bicycle conflict and so on.
15 February 2014
Jing Zhang | Qinke Peng | Shiquan Sun | Che Liu
Abstract: Personalized recommendation is an effective method for fighting “information overload”. However, its performance is often limited by several factors, such as sparsity and cold-start. Some researchers utilize user-created tags of social tagging system to depict user preferences for personalized recommendation, but it is difficult to identify users with similar interests due to the differences between users’ descriptive habits and the diversity of language expression. In order to find a better way to depict user preferences to make it more suitable for personalized recommendation, we introduce a framework that utilizes item domain features to construct user preference models and combines these models with collaborative filtering (CF). The framework not only integrates domain characteristics into a personalized recommendation, but also aids to detecting the implicit relationships among users, which are missed by the conventional CF method. The experimental results show our method achieves the better result, and prove the user preference model is more effective for recommendation.
15 January 2007
M. Nekovee | Y. Moreno | G. Bianconi | M. Marsili
Abstract: We introduce a general stochastic model for the spread of rumours, and derive mean-field equations that describe the dynamics of the model on complex social networks (in particular, those mediated by the Internet). We use analytical and numerical solutions of these equations to examine the threshold behaviour and dynamics of the model on several models of such networks: random graphs, uncorrelated scale-free networks and scale-free networks with assortative degree correlations. We show that in both homogeneous networks and random graphs the model exhibits a critical threshold in the rumour spreading rate below which a rumour cannot propagate in the system. In the case of scale-free networks, on the other hand, this threshold becomes vanishingly small in the limit of infinite system size. We find that the initial rate at which a rumour spreads is much higher in scale-free networks than in random graphs, and that the rate at which the spreading proceeds on scale-free networks is further increased when assortative degree correlations are introduced. The impact of degree correlations on the final fraction of nodes that ever hears a rumour, however, depends on the interplay between network topology and the rumour spreading rate. Our results show that scale-free social networks are prone to the spreading of rumours, just as they are to the spreading of infections. They are relevant to the spreading dynamics of chain emails, viral advertising and large-scale information dissemination algorithms on the Internet.
1 October 2013
Yong Tang | Yong Luo | Jie Xiong | Fei Zhao | Yi-Cheng Zhang
Abstract: The impact of monetary policy changes on the monetary market and stock market in China is investigated in this study. The changes of two major monetary policies, the interest rate and required reserve ratio, are analyzed in a study period covering seven years on the interbank monetary market and Shanghai stock market. We find that the monetary market is related to the macro economy trend and we also find that the monetary change surprises both of lowering and raising bring significant impacts to the two markets and the two markets respond to the changes differently. The results suggest that the impact of fluctuations is much larger for raising policy changes than lowering changes in the monetary market on policy announcing and effective dates. This is consistent with the “sign effect”, i.e. bad news brings a greater impact than good news. By studying the event window of each policy change, we also find that the “sign effect” still exists before and after each change in the monetary market. A relatively larger fluctuation is observed before the event date, which indicates that the monetary market might have a certain ability to predict a potential monetary change, while it is kept secret by the central bank before official announcement. In the stock market, we investigate how the returns and spreads of the Shanghai stock market index respond to the monetary changes. Evidences suggest the stock market is influenced but in a different way than the monetary market. The climbing of returns after the event dates for the lowering policy agrees with the theory that lowering changes can provide a monetary supply to boost the market and drive the stock returns higher but with a delay of 2 to 3 trading days on average. While in the bear market, the lowering policy brings larger volatility to the market on average than the raising ones. These empirical findings are useful for policymakers to understand how monetary policy changes impact the monetary and stock markets especially in an emerging market like China where the economy is booming and the policy changes impact the markets as surprises by the central bank without a pre-decided schedule. This is totally different from previous studies on FED, which follows pre-decided schedules for monetary policy changes.
1 February 2014
Joonhyun Bae | Sangwook Kim
Abstract: Identifying influential spreaders is an important issue in understanding the dynamics of information diffusion in complex networks. The k-shell index, which is the topological location of a node in a network, is a more efficient measure at capturing the spreading ability of a node than are the degree and betweenness centralities. However, the k-shell decomposition fails to yield the monotonic ranking of spreaders because it assigns too many nodes with the same k-shell index. In this paper, we propose a novel measure, coreness centrality, to estimate the spreading influence of a node in a network using the k-shell indices of its neighbors. Our experimental results on both real and artificial networks, compared with an epidemic spreading model, show that the proposed method can quantify the node influence more accurately and provide a more monotonic ranking list than other ranking methods.
1 November 2013
Kaihe Liu | Yanfei Jin
Abstract: We studied the motion of an underdamped Brownian particle in a periodic potential subject to a harmonic excitation and a colored noise. The average input energy per period and the phase lag are calculated to quantify the phenomenon of stochastic resonance (SR). The numerical results show that most of the out-of-phase trajectories make a transition to the in-phase state as the temperature increases. And the colored noise delays the transitions between these two dynamical states. The each curve of the average input energy per period and the phase lag versus the temperature exist a mono peak and SR appears in this system. Moreover, the optimal temperature where the SR occurs becomes larger and the region of SR grows wider as the correlation time of colored noise increases.
1 December 2013
Jiri Kukacka | Jozef Barunik
Abstract: The main aim of this work is to incorporate selected findings from behavioural finance into a Heterogeneous Agent Model using the Brock and Hommes (1998) [34] framework. Behavioural patterns are injected into an asset pricing framework through the so-called ‘Break Point Date’, which allows us to examine their direct impact. In particular, we analyse the dynamics of the model around the behavioural break. Price behaviour of 30 Dow Jones Industrial Average constituents covering five particularly turbulent US stock market periods reveals interesting patterns in this aspect. To replicate it, we apply numerical analysis using the Heterogeneous Agent Model extended with the selected findings from behavioural finance: herding, overconfidence, and market sentiment. We show that these behavioural breaks can be well modelled via the Heterogeneous Agent Model framework and they extend the original model considerably. Various modifications lead to significantly different results and model with behavioural breaks is also able to partially replicate price behaviour found in the data during turbulent stock market periods.
15 December 2002
Jan W. Kantelhardt | Stephan A. Zschiegner | Eva Koscielny-Bunde | Shlomo Havlin | Armin Bunde | H.Eugene Stanley
Abstract: We develop a method for the multifractal characterization of nonstationary time series, which is based on a generalization of the detrended fluctuation analysis (DFA). We relate our multifractal DFA method to the standard partition function-based multifractal formalism, and prove that both approaches are equivalent for stationary signals with compact support. By analyzing several examples we show that the new method can reliably determine the multifractal scaling behavior of time series. By comparing the multifractal DFA results for original series with those for shuffled series we can distinguish multifractality due to long-range correlations from multifractality due to a broad probability density function. We also compare our results with the wavelet transform modulus maxima method, and show that the results are equivalent.
1 March 2014
Bogdan Negrea
Abstract: This paper postulates the concept of financial market energy and provides a statistical measure of the financial market crisis magnitude based on an analogy between earthquakes and market crises. The financial energy released by the market is expressed in terms of trading volume and stock market index returns. A financial “earthquake” occurs if the financial energy released by the market exceeds the estimated threshold of market energy called critical energy. Similar to the Richter scale which is used in seismology in order to measure the magnitude of an earthquake, we propose a financial Gutenberg–Richter relation in order to capture the crisis magnitude and we show that the statistical pattern of the financial market crash is given by two statistical regimes, namely Pareto and Wakeby distributions.
引用排名
Volume 388, Issues 2-3, January 2009, Pages 193-208
Peng, K. | Yang, Y.
In this paper, we study a leader-following consensus problem for a multi-agent system with a varying-velocity leader and time-varying delays. Here, the interaction graph among the followers is switching and balanced. At first, we propose a neighbor-based rule for every agent to track a leader whose states may not be measured. In addition, we consider the convergence analysis of this multi-agent system under two different conditions: the connection between the followers and the leader is time-invariant and time-varying. For the first case, a novel decomposition method is introduced to facilitate the convergence analysis. By utilizing a Lyapunov-Krasovskii functional, we obtain sufficient conditions for uniformly ultimately boundedness of the tracking errors. Finally, two simulations are also presented to illustrate our theoretical results. © 2008 Elsevier B.V. All rights reserved.
Volume 388, Issue 8, April 2009, Pages 1706-1712
Shen, H. | Cheng, X. | Cai, K. | Hu, M.-B.
Clustering and community structure is crucial for many network systems and the related dynamic processes. It has been shown that communities are usually overlapping and hierarchical. However, previous methods investigate these two properties of community structure separately. This paper proposes an algorithm (EAGLE) to detect both the overlapping and hierarchical properties of complex community structure together. This algorithm deals with the set of maximal cliques and adopts an agglomerative framework. The quality function of modularity is extended to evaluate the goodness of a cover. The examples of application to real world networks give excellent results. © 2008 Elsevier B.V. All rights reserved.
Volume 390, Issue 6, March 2011, Pages 1150-1170
Linyuan, L.L. | Zhou, T.
Link prediction in complex networks has attracted increasing attention from both physical and computer science communities. The algorithms can be used to extract missing information, identify spurious interactions, evaluate network evolving mechanisms, and so on. This article summaries recent progress about link prediction algorithms, emphasizing on the contributions from physical perspectives and approaches, such as the random-walk-based methods and the maximum likelihood methods. We also introduce three typical applications: reconstruction of networks, evaluation of network evolving mechanism and classification of partially labeled networks. Finally, we introduce some applications and outline future challenges of link prediction algorithms. © 2010 Elsevier B.V. All rights reserved.
Volume 388, Issue 21, November 2009, Pages 4586-4592
Sun, H. | Chen, W. | Chen, Y.
The purpose of this paper is to offer a unified discussion of variable-order differential operators in anomalous diffusion modeling. The characteristics of the new models, in contrast to constant-order fractional diffusion models, change with time, space, concentration or other independent quantities. We introduced a classification of variable-order fractional diffusion models based on the possible physical origins which prompt the variable-order. Some potential applications of the variable-order fractional diffusion models are also discussed. © 2009 Elsevier B.V. All rights reserved.
Volume 389, Issue 1, January 2010, Pages 179-186
Zhang, Z.-K. | Zhou, T. | Zhang, Y.-C.
Personalized recommender systems are confronting great challenges of accuracy, diversification and novelty, especially when the data set is sparse and lacks accessorial information, such as user profiles, item attributes and explicit ratings. Collaborative tags contain rich information about personalized preferences and item contents, and are therefore potential to help in providing better recommendations. In this article, we propose a recommendation algorithm based on an integrated diffusion on user-item-tag tripartite graphs. We use three benchmark data sets, Del.icio.us, MovieLens and BibSonomy, to evaluate our algorithm. Experimental results demonstrate that the usage of tag information can significantly improve accuracy, diversification and novelty of recommendations. © 2009 Elsevier B.V. All rights reserved.
Volume 388, Issue 8, April 2009, Pages 1571-1576
Sornette, D. | Woodard, R. | Zhou, W.-X.
We present an analysis of oil prices in USD and in other major currencies that diagnoses unsustainable faster-than-exponential behavior. This supports the hypothesis that the recent oil price run-up was amplified by speculative behavior of the type found during a bubble-like expansion. We also attempt to unravel the information hidden in the oil supply-demand data reported by two leading agencies, the US Energy Information Administration (EIA) and the International Energy Agency (IEA). We suggest that the found increasing discrepancy between the EIA and IEA figures provides a measure of the estimation errors. Rather than a clear transition to a supply restricted regime, we interpret the discrepancy between the IEA and EIA as a signature of uncertainty, and there is no better fuel than uncertainty to promote speculation! Our post-crash analysis confirms that the oil peak in July 2008 occurred within the expected 80% confidence interval predicted with data available in our pre-crash analysis. © 2009.
https://wap.sciencenet.cn/blog-329471-772230.html
上一篇：Intensive course on Financial Crises
下一篇：巴拉巴西教授可以正式招收网络科学研究方向的博士生了
