Matus Medo

Bipartite network projection and personal recommendation.

Tao Zhou, Jie Ren, Matúš Medo, and Yi-Cheng Zhang Department of Physics, University of Fribourg, Chemin du Muse 3, CH-1700 Fribourg, Switzerland Department of Modern Physics and Nonlinear Science Center, University of Science and Technology of China, Hefei Anhui, 230026, PR China Information Economy and Internet Research Laboratory, University of Electronic Science and Technology of China, Chengdu Sichuan, 610054, PR China (Dated: February 1, 2008)

Solving the apparent diversity-accuracy dilemma of recommender systems

Recommender systems use data on past user preferences to predict possible future likes and interests. A key challenge is that while the most useful individual recommendations are to be found among diverse niche objects, the most reliably accurate results are obtained by methods that recommend objects based on user or object similarity. In this paper we introduce a new algorithm specifically to address the challenge of diversity and show how it can be used to resolve this apparent dilemma when combined in an elegant hybrid with an accuracy-focused algorithm. By tuning the hybrid appropriately we are able to obtain, without relying on any semantic or context-specific information, simultaneous gains in both accuracy and diversity of recommendations.

Temporal Effects in the Growth of Networks

We show that to explain the growth of the citation network by preferential attachment (PA), one has to accept that individual nodes exhibit heterogeneous fitness values that decay with time. While previous PA-based models assumed either heterogeneity or decay in isolation, we propose a simple analytically treatable model that combines these two factors. Depending on the input assumptions, the resulting degree distribution shows an exponential, log-normal or power-law decay, which makes the model an apt candidate for modeling a wide range of real systems.

Emergence of Scale-Free Leadership Structure in Social Recommender Systems

The study of the organization of social networks is important for the understanding of opinion formation, rumor spreading, and the emergence of trends and fashion. This paper reports empirical analysis of networks extracted from four leading sites with social functionality (Delicious, Flickr, Twitter and YouTube) and shows that they all display a scale-free leadership structure. To reproduce this feature, we propose an adaptive network model driven by social recommending. Artificial agent-based simulations of this model highlight a “good get richer” mechanism where users with broad interests and good judgments are likely to become popular leaders for the others. Simulations also indicate that the studied social recommendation mechanism can gradually improve the user experience by adapting to tastes of its users. Finally we outline implications for real online resource-sharing systems.

Information filtering by similarity-preferential diffusion processes

Recommender systems provide a promising way to address the information overload problem which is common in online systems. Based on past user preferences, a recommender system can find items that are likely to be relevant to a given user. Two classical physical processes, mass diffusion and heat conduction, have been used to design recommendation algorithms and a hybrid process based on them has been shown to provide accurate and diverse recommendation results. We modify both processes as well as their hybrid by introducing a parameter which can be used to enhance or suppress the weight of users who are most similar to the target user for whom the recommendation is done. Evaluation on two benchmark data sets demonstrates that both recommendation accuracy and diversity are improved for a wide range of parameter values. Threefold validation indicates that the achieved results are robust and the new recommendation methods are thus applicable in practice.

MET Inhibition Results in DNA Breaks and Synergistically Sensitizes Tumor Cells to DNA-Damaging Agents Potentially by Breaching a Damage-Induced Checkpoint Arrest.

While recent studies implicate that signaling through the receptor tyrosine kinase MET protects cancer cells from DNA damage, molecular events linking MET to the DNA damage response machinery are largely unknown. Here, we studied the impact of MET inhibition by the small molecule PHA665752 on cytotoxicity induced by DNA-damaging agents. We demonstrate that PHA665752 reduces clonogenic survival of tumor cells with MET overexpression when combined with ionizing radiation and synergistically cooperates with ionizing radiation or adriamycin to induce apoptosis. In search of mechanisms underlying the observed synergism, we show that PHA665752 alone considerably increases γH2AX levels, indicating the accumulation of double-strand DNA breaks. In addition, PHA665752 treatment results in sustained high levels of γH2AX and phosphorylated ATM postirradiation, strengthening the assumption that MET inhibition attenuates postdamage DNA repair. PHA665752, alone or in combination with irradiation, leads also to a massive increase of γH2AX tyrosine phosphorylation and its subsequent interaction with the proapoptotic kinase JNK1. Finally, MET inhibition reduces activation of ATR, CHK1, and CDC25B and abrogates an associated DNA damage-induced S phase arrest. This indicates that MET inhibition compromises a critical damage-dependent checkpoint that may enable DNA-damaged cells to exit cell cycle arrest before repair is completed.

Ranking in evolving complex networks

Complex networks have emerged as a simple yet powerful framework to represent and analyze a wide range of complex systems. The problem of ranking the nodes and the edges in complex networks is critical for a broad range of real-world problems because it affects how we access online information and products, how success and talent are evaluated in human activities, and how scarce resources are allocated by companies and policymakers, among others. This calls for a deep understanding of how existing ranking algorithms perform, and which are their possible biases that may impair their effectiveness. Well-established ranking algorithms (such as the popular Googles PageRank) are static in nature and, as a consequence, they exhibit important shortcomings when applied to real networks that rapidly evolve in time. The recent advances in the understanding and modeling of evolving networks have enabled the development of a wide and diverse range of ranking algorithms that take the temporal dimension into account. The aim of this review is to survey the existing ranking algorithms, both static and time-aware, and their applications to evolving networks. We emphasize both the impact of network evolution on well-established static algorithms and the benefits from including the temporal dimension for tasks such as prediction of real network traffic, prediction of future links, and identification of highly-significant nodes.

Network-based recommendation algorithms: A review

Recommender systems are a vital tool that helps us to overcome the information overload problem. They are being used by most e-commerce web sites and attract the interest of a broad scientific community. A recommender system uses data on users’ past preferences to choose new items that might be appreciated by a given individual user. While many approaches to recommendation exist, the approach based on a network representation of the input data has gained considerable attention in the past. We review here a broad range of network-based recommendation algorithms and for the first time compare their performance on three distinct real datasets. We present recommendation topics that go beyond the mere question of which algorithm to use–such as the possible influence of recommendation on the evolution of systems that use it–and finally discuss open research directions and challenges.

Heterogeneity, quality, and reputation in an adaptive recommendation model

Abstract. Recommender systems help people cope with the problem of information overload. A recently proposed adaptive news recommender model [M. Medo, Y.-C. Zhang, T. Zhou, Europhys. Lett. 88, 38005 (2009)] is based on epidemic-like spreading of news in a social network. By means of agent-based simulations we study a “good get richer” feature of the model and determine which attributes are necessary for a user to play a leading role in the network. We further investigate the filtering efficiency of the model as well as its robustness against malicious and spamming behaviour. We show that incorporating user reputation in the recommendation process can substantially improve the outcome.

TREND PREDICTION IN TEMPORAL BIPARTITE NETWORKS: THE CASE OF MOVIELENS, NETFLIX, AND DIGG

Online systems, where users purchase or collect items of some kind, can be effectively represented by temporal bipartite networks where both nodes and links are added with time. We use this representation to predict which items might become popular in the near future. Various prediction methods are evaluated on three distinct datasets originating from popular online services (Movielens, Netflix, and Digg). We show that the prediction performance can be further enhanced if the user social network is known and centrality of individual users in this network is used to weight their actions.