报告人：Matus Medo

报告题目：复杂信息网络的模型与算法

Models and algorithms for complex

Information networks

报告时间：2018年4月13日上午10:30

报告地点：北翼楼4楼多媒体室

报告摘要：

Complex networks have been successfully used to represent a wide range of real systems. We will focus on complex information networks in particular where growth and time play crucial roles and which are relevant in diverse fields such as bibliometrics (to model citations among scientific papers), e-commerce (to model users and their purchases on Amazon.com, for example), and information filtering in general (to model links among web sites, for example). We will first show that to model these systems, the classical preferential attachment model needs to be augmented by introducing node fitness and aging. We will use the model to demonstrate that usual network metrics and algorithms that ignore the time information produce inferior or even misleading results. The identified problems directly motivate the development of new metrics and algorithms. We will thus conclude by showing how taking time into account leads to better ranking and recommendation in complex networks.

报告人简介：

Matus Medo男，博士（后），瑞士弗里堡大学副教授，电子科技大学百人计划研究员。长期从事统计物理，特别是复杂系统与复杂网络等方向的研究工作，在复杂网络模型、信息推荐系统、博弈论、经济物理等相关领域取得了较为突出的学术成果。已经发表学术论文49篇，其中第一作者或通讯作者论文20余篇，发表期刊包括Physical Review Letters、Physics Reports、Proceedings of the National Academy of Sciences of the United States of America等国际知名期刊，Google Scholar统计的引用近2000次，单篇最高引用超650次；SCI他引近1000次，其中3篇论文他引次数超过200次，入选ESI高引用论文，被包括Science、PNAS等期刊在内的知名杂志引用。研究成果得到了复杂网络研究领域权威、哈佛大学A. Barabási教授、美科院院士H. Eugene Stanley教授等学者的肯定。主持1项瑞士自然科学基金，作为骨干成员完成2项瑞士自然科学基金和4项欧盟基金，总经费超过1000万元人民币。

欢迎老师和同学踊跃参加！