Physical approach to complex systems

doi:10.1016/j.physrep.2012.01.007

Physical approach to complex systems

2012-06-01

383

PRE

AI

DOI：10.1016/j.physrep.2012.01.007

原文链接

原文求助

摘要

En 中文

Typically, complex systems are natural or social systems which consist of a large number of nonlinearly interacting elements. These systems are open, they interchange information or mass with environment and constantly modify their internal structure and patterns of activity in the process of self-organization. As a result, they are flexible and easily adapt to variable external conditions. However, the most striking property of such systems is the existence of emergent phenomena which cannot be simply derived or predicted solely from the knowledge of the systems' structure and the interactions among their individual elements. This property points to the holistic approaches which require giving parallel descriptions of the same system on different levels of its organization. There is strong evidence - consolidated also in the present review - that different, even apparently disparate complex systems can have astonishingly similar characteristics both in their structure and in their behaviour. One can thus expect the existence of some common, universal laws that govern their properties. Physics methodology proves helpful in addressing many of the related issues. In this review, we advocate some of the computational methods which in our opinion are especially fruitful in extracting information on selected - but at the same time most representative - complex systems like human brain, financial markets and natural language, from the time series representing the observables associated with these systems. The properties we focus on comprise the collective effects and their coexistence with noise, long-range interactions, the interplay between determinism and flexibility in evolution, scale invariance, criticality, multifractality and hierarchical structure. The methods described either originate from hard physics - like the random matrix theory - and then were transmitted to other fields of science via the field of complex systems research, or they originated elsewhere but turned out to be very useful a so in physics like, for example, fractal geometry. Further methods discussed borrow from the formalism of complex networks, from the theory of critical phenomena and from nonextensive statistical mechanics. Each of these methods is helpful in analyses of specific aspects of complexity and all of them are mutually complementary. (C) 2012 Elsevier B.V. All rights reserved.

Keyword:

Complex systems

Complexity measures

Correlations

Asymmetric correlations

Coexistence of collectivity and noise

Random matrix theory

Time series analysis

Fractals

Multifractals

Critical phenomena

Complex networks

Financial markets

Human brain

Quantitative linguistics

Physical approach to complex systems

摘要

AI总结

期刊

机构

引用论文