Post on 06-May-2015
Embed Size (px)
DESCRIPTIONAn exploration of complex network theory and its potential uses for futures. A presentation to the Association of Professional Futurists.
- 1.Possibility NetworksAn Exploration of Complex Network Theory and Its Potential Uses for FuturesFor the Association of Professional Futurists Professional DevelopmentSeminarChicago, Illinois July 29th, 2005 By: David A. Jarvis
2. An Opening Thought The greatest challenge today, not just in cellbiology and ecology but in all of science, is the accurate and complete description of complexsystems. Scientists have broken down manykinds of systems. They think they know most ofthe elements and forces. The next task is to reassemble them, at least in mathematicalmodels that capture the key properties of the entire ensembles. - E.O. Wilson, Consilience: The Unity of Knowledge 2 3. Why Complex Networks and Futures? It has been expressed by members of the APF that the futures fieldneeds new tools, techniques and methodologies The fields last major development was scenario planning, whichevolved from military planning during World War II and was adoptedby the corporate world in the 1960s In a recent APF professional development survey, members saidthey wanted more information on simulation and games, chaos andagent-based models Complex systems can significantly augment the spectrum of tools thatfuturists can offer clients and organizations Those trained in studying the future have explored systems thinking,chaos and complex systems, but tools and applications have not widelymoved beyond the metaphorical level Where do complex systems and the systemic study of the futureintersect? Can new tools be created for futurists extracted from theresearch done in complex systems? 3 4. What to Expect Gain a basic understanding of the science and mathbehind network theory - what it is and what it isnt Learn about the major players in network theory and thefoundational books and papers for the field Understand the theoretical basis behind such concepts asdiffusion of innovations and idea contagions Learn how social networks can be used as a futures tool Participate in an exercise demonstrating the usefulness andpower of social networks This is a BROAD and SHALLOW view of network theory its purpose is to stimulate thinking and help form questions4 5. IntroductionI. History and BackgroundII. Scientific BasicsIII. Examples and ApplicationsIV. Demonstration5 6. Definition of a Complex Network A society tends to view itself through a lens ofthe technologies it creates Networks are EVERYWHERE! Power grids Computer networks Ecological systems (e.g. food webs) Social interaction patterns Romantic and sexual networks The Internet and World Wide Web Transportation (roads, airlines, rail, etc.) Communication networks (phones, post, etc.) Protein interactions and cellular networks Biological systems (the brain, circulatory system, etc.) 6 7. Definition of a Complex Network Complex system - a collection of interacting elements arranged forpurpose that exhibits high-dimensionality, non-linearity, sensitivedependence of initial conditions, and possibly emergent behavior Complex network - a representation of a complex system, comprisedof nodes and links 7 8. I. History and Background 8 9. The Seven Bridges of Knigsberg Question: Is it possible to cross all seven bridgesonly once and return to your starting point? In 1736, Leonhard Euler proved that it was notpossible through one of the first formalmathematical discussions using graph theoryNodeLink 9 10. Paul Erds Hungarian mathematician andprolific scientific author With Alfrd Rnyi did fundamentalresearch into how networks form Discovered random networktheory simplest method ofcreating a network, God plays dice Emergence of a giant component Erds number small worldphenomenon10 11. Buttons and Strings11 12. The Strength of Weak Ties Mathematical sociologist Mark Granovetter (article inAmerican Journal of Sociology, 1973) the degree of overlap of two individuals friendshipnetworks varies directly with the strength of their tie toone another. Weak ties can serve as bridges between different socialgroups, allow you to reach more people more quickly Strong ties lead tofragmentation, weak ties leadto integration Example: finding a job12 13. Six Degrees of Separation Hungarian author Karinthys short story entitledChains (1929) Milgrams experiment (1967) Find the distance between any two people in the U.S. Sent a letter to a few hundred randomly selectedpeople from Boston and Omaha with instructions tosend to a Massachusetts stockbroker, the recipientcould only send the letter to someone they knew on afirst name basis Common sense says it should take hundreds of steps,it only took six on average, its a small world after all! Idealized vs. real social networks13 14. Small Worlds & Scale Free Small world networks Duncan Watts and Steven Strogatz (1998) Each node can reach every other node in a small numberof steps Characterized by high clustering, short characteristic pathlengths Scale-free networks Albert-Lszl Barabsi (professor of physics at NotreDame) & Rka Albert (currently at Penn State) Examined networks that exhibited a power-law distributionin their degree (Internet and WWW) Large number of poorly connected nodes and a small number of well-connected hubs14 15. Scale-Free Networks Poisson distributions vs. power-law distributions Power law example: distribution of wealthNumber of nodes with k linksNumber of nodes with k linksMost nodeshave the sameLarge numbernumber of linksof nodes have few linksSmall number ofnodes (hubs) havemany linksNumber of links (k) Number of links (k) Normal (Poisson) DistributionPower-Law DistributionAdapted from: Linked, Barabsi , pg. 7115 16. Related Topics Fads Memes Chaos Theory Social Networking Diffusion of Innovations Contagion Agent-Based Modeling Collective Robotics and Distributed Systems Emergence 16 17. Fads Definition Ideas or things in a culture that becomeextremely popular very quickly, and just as quicklybecome unpopular; linked to herd mentality Bandwagon effect a benefit that a person enjoys as a result of others doing the same thing that they do Relation accelerated s-curve behavior Examples Irrational exuberance in the stock market Flash mobs Christmas toys Fashion Music & dance crazes 17 18. Memes Definition concept created by Richard Dawkins inhis book The Selfish Gene (1976); a piece ofinformation that can be transmitted between twominds; parallels to evolution Relation Alternate explanationfor how ideas propagatethrough a society18 19. Chaos Theory Definition The irregular, unpredictable behaviorof deterministic, nonlinear dynamical systems.,Roderick V. Jensen, Yale University Relation descriptive of natural systems, sensitivityto initial conditions, patterns Examples Double pendulums Multi-body gravitational problems Turbulent fluids (e.g. the atmosphere) Work of Lorenz, Mandelbrot19 20. Diffusion of Innovations Definition The theories of diffusion can trace their roots back to the French sociologist Gabriel Tarde who identified the innovation adoption S- curve, group mind, laws of imitation Progressed through the agricultural research of Ryan and Gross in the 1940s, lead to the notion of adopter categories (innovators, early adopters, early majority, late majority, laggards) Rogers seminal work Diffusion of Innovation (1962) formalized these theories Relation there are many researches who study thediffusion of innovations in complex networks Information Flow in Social Groups A generalized model of social and biological contagion Modeling diffusion of innovations in a social network The Power of a Good Idea: Quantitative Modeling of the Spread of Ideas from Epidemiological Models 20 21. ContagionVIDEOContact Networks in Predicting and Controlling Emerging Infectious DiseasesLauren Ancel MeyersSFI External Faculty, University of Texas at Austin7:00 - 15:30 Background30:00 - 36:00 Contact Network Epidemiology21 22. Agent-Based Modeling Definition ABM is a simulation tool that ischaracterized by large numbers of simple agentsinteracting through well defined rule sets Relation ABM is widely used as a tool formodeling complex adaptive systems Examples Crowd dynamics Traffic patterns Economic markets Insect behavior Genetic algorithmsBonabeau, Eric (2002) Proc. Natl. Acad. Sci. USA 99, 7280-7287 22 23. Emergence Definition surprising or unexpected global results that canoccur when the parts of any system interact locally viasimple rules; the whole is greater than the sum of its parts Relation Emergent behavior arises in complex systems;self-organization Examples Human consciousness Traffic patterns Galaxy formation Ant colonies, flocking behavior Urban evolution Life 23 24. Emergence I begin to think that this matter of late emergentproperties that the physicists talk about when theydiscuss complexity and cascading sensitivities is an important concept for historians. Justice may be anlate emergent property. And maybe we can glimpse the beginnings of it emerging; or maybe it emergedlong ago, among the primates and proto-humans,and is only now gaining leverage in the world, aided by the material possibility of postscarcity. - The Years of Rice and Salt, Kim Stanley Robinson 24 25. Social Networking Definition businessand social networkingservices Relation uses complexnetwork principles likesmall worlds and sixdegrees of separation Examples Friendster LinkedIn Orkut Yahoo 360 MySpace Ryze25 26. Collective Robotics Definition large numbers of coordinated simple robotsdesigned to perform a complex task, inspired by socialinsects Relation still a very immature technology, collectiverobotics uses agent-based modeling and principles ofemergence Examples Swarms of unmanned military vehicles (air, land, sea) Mobile sensors networks for ocean research, search and rescue, etc.Image taken from iRobot website26 27. Complex Network Literature Multi-disciplinary Most works are fairly recent Still no definitive academic textbook oncomplex networks Three levels Metaphor Popu