Abstract
Network analysis views complex systems as networks with well-defined structural properties that account for their complexity. These characteristics, which include scale-free behavior, small worlds and communities, are not to be found in networks such as random graphs and lattices that do not correspond to complex systems. They provide therefore a robust ground for claiming the existence of “complex networks” as a non-trivial subset of networks. The theory of complex networks has thus been successful in making systematically explicit relevant marks of complexity in the form of structural properties, and this success is at the root of its current popularity. Much less systematic has been, on the other hand, the definition of the properties of the building components of complex networks. The obvious assumption is that these components must be nodes and links. Generally, however, the internal structure of nodes is not taken into account, and links are serendipitously identified by the perspective with which one looks at the network to be analyzed. For instance, if the nodes are Web pages that contain information about scientific papers, one point of view will match the relevant links with hyperlinks to similar Web pages, and another with citations of other articles. We intend to contribute here a systematic approach to the identification of the components of a complex network that is based on information theory. The approach hinges on some recent results arising from the convergence between the theory of complex networks and probabilistic techniques for content mining. At its core there is the idea that nodes in a complex network correspond to basic information units from which links are extracted via methods of machine learning. Hence the links themselves are viewed as emergent properties, similarly to the broader structural properties mentioned above. Indeed, beside rounding up the theory, this approach based on learning has clear practical benefits, in that it makes networks emerge from arbitrary information domains. We provide examples and applications in a variety of contexts, starting from an information-theoretic reconstruction of the well-known distinction between “strong links” and “weak links” and then delving into specific applications such as business process management and analysis of policy making.
Highlights
The study of complex systems has recently received a vigorous surge from the evidence that, in addition to those among such systems that have evolved in nature as well as in society during a more or less remote past, many of the most relevant developments made possible by the coming of age of a digital society bear the marks of complexity
The intent of the paper is conceptual and methodological, and relies on the technical background and the formal results of Rossetti et al (2014). It is structured for its remaining parts as follows: in Sect. 2 we provide an information-theoretic framework for complex networks; in Sect. 3 we demonstrate how such framework successfully fulfills the objective of making complexity phenomena systematically emerge from an information substratum in the form of the well-known structural properties studied in network theory, and we especially focus on properties of special interest for social network analysis such as communities and the distinction between weak links and strong links; in Sect. 4 we go into specific case studies and applications in such domains as business process management and policy analysis
The topics are acquired via Relational Topic Models (RTM) (Chang and Blei 2009), a probabilistic topic modeling technique, which is elsewhere exploited to capture coupling among classes in object-oriented software systems (Gethers and Poshyvanyk 2010)
Summary
The study of complex systems has recently received a vigorous surge from the evidence that, in addition to those among such systems that have evolved in nature as well as in society during a more or less remote past, many of the most relevant developments made possible by the coming of age of a digital society bear the marks of complexity. Scientific articles, one point of view will match the relevant links with hyperlinks to similar Web pages, and another with citations of other articles Theoretical incompleteness aside, this state of affairs severely limits the practical applicability of network analysis to real cases of complexity, because of the arbitrariness in setting out the initial premises that it entails, and the reservations that inevitably ensue on the actual empirical validity and effective significance of the results. As simple and straightforward as it is, this premise is the keystone of all, since it provides a stringent motivation for an information-driven overhaul of the state of the art of complex networks as well as points the way for its satisfactory implementation The motivation stems both from the opportunity of automating and from the necessity of making less arbitrary the identification of the links that boots up network analysis. The road to a satisfactory implementation pursues the possibility to apply techniques of text analysis and machine learning to the nodes of the network, that are viewed as texts, so as to automatically generate and extract links through an empirically grounded procedure
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
