Network Science Literacy An easy introduction to Systems

Network Science Literacy: An easy introduction to Systems Thinking Mark Sellers, Systems Engineer

Acknowledgements All of the systems pioneers that have felt the itch that something was incomplete in our understanding of the world. Dr. Hiroki Sayama Director, Center for Collective Dynamics of Complex Systems (Co. Co) Director, Graduate Programs in Systems Science Thomas J. Watson School of Engineering and Applied Science, Binghamton University, State University of New York 2

Brief History of Systems before Network Science

‘Conventional Thinking’ & ‘Systems Thinking’ Conventional thinking: Science 300 years in the making: Newton, Galileo and many others Precise, powerful conceptual tool that reduces complex problems to isolated variables Avoids things too wide open… Physics keeps distance from ‘life’ question Social Science, Political Science, Systems Science, etc. : That’s not Science! A dramatically successful tool!: Has driven the growth and extent of our global society Systems Thinking Addresses holistic problems that reductionist science cannot Cybernetics, Chaos, Emergence, Systems Engineering, non-linear dynamics, Self-Organization, Networks, etc. Where are you? Systems thinking is not a new concept to any problem solver. Successful businesses take structure and processes as seriously as the product line. Seasoned engineers consider environment, and ‘illities’ as seriously as the components. Home purchase decisions include schools, traffic, taxes, utilities and more. The short-term successes of ‘Scientific Thinking’ 4 ‘Systems Thinking’. impede the transition to

A common sense definition of ‘Systems’ Are systems real? : Ask the oracle – Google (9/9/16) System: About 4. 4 B results Science: About 2. 2 B results Sex: About 3. 8 B results Some simple examples: Economy, Educational system, Government, Evolution, Battlefield, Man-to-man Defense, Traffic, Weather, cruise-control, Card-counting Blackjack, etc. Set theory: S = {T, R} T=Things, R=Relationships among ‘Things’ Your Discipline is generally about ‘T’, Systems Thinking is about ‘R’ ‘System’ is difficult to define conventionally – 5 but we know they are real and defined by relationships!

History of systems up to Network Science “The whole is more than the sum of its parts. ” Aristotle, 350 BC Charles von Ehrenfels 1890, Gestalt Jan Smuts, 1926, “Holism and Evolution” Warren Weaver, 1948, “Science and Complexity” Organized complexity Ludwig von Bertalanffy, 1968, “General System Theory” The computer: The laboratory for Systems Science is impossible without advancing computer technology Today we do have some ‘systems’ focus and problems: Operations Research, Systems Engineering, Evolution, Climate… Obesity, poverty, kudzu, leaded gas, ISIS, business failures of all etc. awareness of systems and systems-thinking failures Ourkinds, growing 6 along with computer technology creates a tinder-box

The Spark: Small-world, and scale-free networks 2 papers: "Collective dynamics of 'smallworld' networks". Watts, D. J. ; Strogatz, S. H. (1998). "Emergence of scaling in random networks". Barabási, Albert-László; Albert, Réka (1999). • • • Provide repeatable, testable network architectures that are not entirely random Cited over 55, 000 times Previous network topology standard was Erdos-Rényi 1959 random networks “A key discovery of network science is that the architecture of networks emerging in various domains of science, nature, and technology are similar to each other, a consequence of being governed by the same organizing principles. Consequently we can use a common set of mathematical tools to explore these systems. ” 7 - Albert-László Barabási

NETWORK LITERACY “Map of Protein Interactions in Yeast” Hawoong Jeong, Korea Advanced Institute of Science and Technology Essential Concepts and Core Ideas http: //sites. google. com/a/binghamton. edu/netscied

Networks are Everywhere Networks describe how things are connected Technical: Communication systems, semantic systems, the Internet, electrical grids, the water supply People: families and friends, e-mail/text exchanges, Facebook/Twitter/Instagram, professional groups Economic products, financial transactions, corporate partnerships, international trades. Biological and ecological: food webs, gene/protein interactions, neuronal networks, pathways of disease spreading Cultural networks—e. g. , Language/literature/art, historical events, people connected to events Networks can exist at various spatial and/or temporal scales. 9 Map of the Internet Showing Major Internet Service Providers Bill Cheswick

Network literacy: Connection and Interaction Utilizes the language and definitions in ‘Graph theory” Leonard Euler: “The Seven Bridges of Königsburg” 1736 By Bogdan Giuşcă - Public domain (PD), based on the image, CC BY-SA 3. 0, https: //commons. wikimedia. org/w/index. php? curid=112920 “Good Will Hunting” puzzles Some definitions: Nodes, vertices, or actors: Entities “Good Will Hunting”, Gus Van Sant, MIRAMAX, 1997 Edges, links, or ties: Connections can be Undirected (symmetric) or Directed Path: A sequence of edges that leads from one node, through other nodes, to another node Degree: How many edges a given node has 10

Network literacy: Revealed Patterns 1 st : Define a network by describing its parts and how they are connected to each other. The properties in a network that you can study include: How degrees are distributed across nodes Which nodes or edges are most important Strengths and/or weaknesses of the network Sub-structure or hierarchy Path length Hubs: Nodes with large degrees. Clusters or communities: group of nodes that are better connected than by chance alone. With these patterns we can: infer how a network was formed make predictions about dynamical processes. make predictions about its future structure. “Think Locally, Act Locally: The Detection of Small, Medium-Sized, and Large Communities in Large Networks”, Physical Review E, Vol. 91, No. 1: 012821. 11 L. G. S. Jeub, K. Balachandran, M. A. Porter, P. J. Mucha, and M. W. Mahoney [2015]

Network literacy: Visualization Communicates ideas in an intuitive, non-technical way. Draw a network by connecting nodes to each other using edges. By hand, like this: Or, there a variety of tools available for visualizing networks. Gephi, Pajek, Network. X, etc. Effective visualization needs creative information design WARNING: Be careful when evaluating visualizations They typically do not tell the whole story 12 Topology of the MBONE, the Internet’s Broadcast Backbone, Which Transmits Real-time Audio and Video Across the Globe

Network literacy: Computer Models of Networks We can study networks using computer technology especially important for large ones with rich structure. Computers can simulate networks with their dynamical processes Free software tools available for network visualization and analysis. Gephi, Pajek, Network. X, etc. Anyone (not just scientists) can construct, visualize, and analyze networks. The Internet is loaded with interesting network data sets. Computer literacy enables network literacy Scientist, data analyst, software engineer, educator, web developer, media creator, etc. 13 Flow of Email Among a Large Project Team. Each Node Represents a Person Color Coded by Department. Valdis Krebs

Network literacy: Comparing Network Systems Examine their similarities and differences. Network properties appear in many seemingly unrelated systems. Implies general principles that apply to multiple domains. Different properties in different systems. Classify networks in different families Gain insight into why they are different Science is typically conducted in separate disciplines. Networks can help to cross disciplinary boundaries Holistic and more complete understanding of the world Networks help transfer knowledge across different areas of study. Router Pinball Graphic Copyright, New York Hall of Science 2004 14

Network literacy: Structure and state Network structure: how parts are connected in a network. Network state: properties of a network’s nodes and edges. Network structure and state can each change over time. The time scales can be either similar or different. Network structure can influence changes of network state. Spread of diseases, behaviors, or memes in a social network, and traffic patterns on the road network in a city. Network state can influence changes of network structure. New “following” edges in social media and new roads to address traffic jams Binary-State Generative Network Automata Simulation 15 Hiroki Sayama, Binghamton University

Example Networks (Eye Candy!)

The network of global corporate control 147 companies (in red) control 80% of the worlds wealth Nodes: Corporations Edges: Ownership (stock, etc Stefania Vitali, James B Glattfelder and Stefano Battiston “The network of global corporate control” Plo. S one, 2011 17

Interlocking Directorates Directors sit on many different corporate boards Nodes: Corporations Location: Corp headquarters Edges: Shared directors The rise of the European corporate elite: evidence from the network of interlocking directorates in 2005 and 2010: Economy and Society: Vol 42, No 1 18

Afghanistan Stability Nodes: beliefs, concepts, Edges: “Influences” “If what I’m hearing is right, there is nothing I can do to solve this. ” 19 Billy Bob Thornton as Brigadier General Hollanek in “Whisky Tango Foxtrot”

Information flow structure in product development Nodes: Design Elements Edges: Shared interface Braha, D. & Bar-Yam, Y. J Inf 20 Technol (2004) 19: 244. doi: 10. 1057/palgrave. jit. 2000030

Information flow in INCOSE SE Processes Nodes: INCOSE SE Processes Edges: Information (inputs & Outputs) Can you see a waterfall in this process mapping? Node Size: # of inputs In-Degree Label size: # of outputs Out-Degree Node color: Betweeness Centrality Likelihood node is ‘between’ 2 other nodes 21

Nodes: Diseases & Genes Edges: Implicated 22

Network of Top Tennis Players Radicchi, 2011 23 Nodes: Players Edges: Defeated by

Student contacts: Temporal relationships Nodes: students Edges: near Columns: 20 sec interval 24

ISIS web sites Nodes: ISIS websites Edges: link Bar graph: High activity 25

Likelihood of crossing the isle in congress. . Nodes: Congressmen Edges: Voted together Color: Party 26

High power Summary Network Science is here to stay! Network science is the enabling micro/telescope of systems science. (system-o-scope? ) Network science is an easy introduction to systems thinking. “It must, in all justice, be admitted that never again will scientific life be as satisfying and serene as in days when determinism reigned supreme. In partial recompense for the tears we must shed and the toil we must endure is the satisfaction of knowing that we are treating significant problems in a more realistic and productive fashion. ” - Richard Bellman 1961 27

Questions? Thank-you 6/3/2021 M W Sellers, Systems Science, SUNY Binghamton 28

Abstract Systems thinking is not a new concept to any problem solver. We are surrounded by examples of successful everyday systems thinking. Successful businesses take structure and processes as seriously as the product line. Seasoned engineers consider environment, and ‘illities’ as seriously as the components. There also many examples of failures in systems thinking: The home mortgage crisis, kudzu, leaded gas, ISIS, business failures of all kinds, etc. The intent of this paper is to begin to provide a common language for thinking about systems across all groups and disciplines within NG using the emerging discipline: Network Science. Galileo, Newton and others, beginning over 400 years ago, described a reductionist and powerful conceptual tool that has altered the path of our global society. The success of this simple concept for day-to-day problems has impeded the transition to systems thinking. New theories like Cybernetics, Chaos, Fuzzy logic, Uncertainty, non-linear dynamics, Self-Organization and others at best are only incorporated as fringe elements in our day-to-day lives. However, combined with our growing understanding of systems-thinking failures, these theories created a tinder-box of ideas that only needed a spark. That spark came in the form of 2 independent papers published in 1998 & 1999: Watts, D. J. ; Strogatz, S. H. (1998). "Collective dynamics of 'small-world' networks". Barabási, Albert-László; Albert, Réka (1999). "Emergence of scaling in random networks". These papers deal with important concepts that can only be understood as complex systems – the components are not important. Together these 2 papers have been cited over 55, 000 times. “Network Science” is now used for economics, disease, telecom, genetics, evolution and much more. Network Science is arguably leading the global shift in systems-thinking. Network Science uses an easy, systems theory and the graphically presented networks are powerful intuitive teaching tools. Network topologies used in this paper: *Random graph (probability) *Small-world model (6 deg of separation) *Scale-free model (why the internet never crashes and computer viruses will never go away) This paper will introduce these useful models and others in the context of current systems thinking to explain the emerging descriptive vocabulary and metrics: Nodes, edges, communities, degree, clustering, centrality, dynamic networks, dynamics on networks, robustness and more. Many well documented networks illustrate Network Science for both a technical and non-technical audience. 29