AI Class Topic 5: Social Network Graph
Download as PPTX, PDF1 like205 views
This document discusses the use of social network graphs and analytics. It provides an overview of key concepts in social network analysis (SNA) including representing social networks, identifying strong and weak ties, central nodes, and overall network structure. Examples are given of how SNA is used in business, law enforcement, social media sites, and more. Key measures discussed include degree, betweenness, closeness, eigenvector centrality, density, and clustering coefficient. The small-world phenomenon and preferential attachment are also covered.
AI Class Topic 5: Social Network Graph
- 1. WELCOME TO: Use Of Social Network Graph And Analytics By Adj Prof. Giuseppe Mascarella [email protected]
- 2. Social Media Innovation Lecture 6
- 3. Social Media Innovation Lecture 9 3 https://www.cmswire.com/digital- workplace/the-enterprise-social- network-graph-battle-who-is-poised-to- win/ In the consumer context, the social graph provides connections with both people and artefacts. Each connection provides a means for personalizing content and ultimately advertisements. In the enterprise SNA context, links to artefacts are used to infer indirect (weaker) connections (e.g. How Microsoft Delve infers worker associations). Direct connections can be both one-way and reciprocated two- way (strongest). It is the people to people graph that provides the intelligence for Enterprise value capture.
- 5. 5 Edge or Original content in this presentation is licensed under the Creative Commons Singapore Attribution 3.0 license unless stated otherwise (see above)
- 6. Social Media Innovation Lecture 6 6
- 7. Social Media Innovation Lecture 6 7
- 8. CNM Social Media Module – Giorgos Cheliotis ([email protected]) 8 Newman et al, 2006 Newman et al, 2006 A very early example of network analysis comes from the city of Königsberg (now Kaliningrad). Famous mathematician Leonard Euler used a graph to prove that there is no path that crosses each of the city’s bridges only once (Newman et al, 2006). SNA has its origins in both social science and in the broader fields of network analysis and graph theory Network analysis concerns itself with the formulation and solution of problems that have a network structure; such structure is usually captured in a graph (see the circled structure to the right) Graph theory provides a set of abstract concepts and methods for the analysis of graphs. These, in combination with other analytical tools and with methods developed specifically for the visualization and analysis of social (and other) networks, form the basis of what we call SNA methods. But SNA is not just a methodology; it is a unique perspective on how society functions. Instead of focusing on individuals and their attributes, or on macroscopic social structures, it centers on relations between individuals, groups, or social institutions
- 9. This is an early depiction of what we call an ‘ego’ network, i.e. a personal network. The graphic depicts varying tie strengths via concentric circles (Wellman, 1998) • CNM Social Media Module – Giorgos Cheliotis ([email protected])9 Wellman, 1998 Studying society from a network perspective is to study individuals as embedded in a network of relations and seek explanations for social behavior in the structure of these networks rather than in the individuals alone. This ‘network perspective’ becomes increasingly relevant in a society that Manuel Castells has dubbed the network society. SNA has a long history in social science, although much of the work in advancing its methods has also come from mathematicians, physicists, biologists and computer scientists (because they too study networks of different types) The idea that networks of relations are important in social science is not new, but widespread availability of data and advances in computing and methodology have made it much easier now to apply SNA to a range of problems
- 10. 10 These visualizations depict the flow of communications in an organization before and after the introduction of a content management system (Garton et al, 1997) A visualization of US bloggers shows clearly how they tend to link predominantly to blogs supporting the same party, forming two distinct clusters (Adamic and Glance, 2005)
- 11. 11 Businesses use SNA to analyze and improve communication flow in their organization, or with their networks of partners and customers 1. Law enforcement agencies (and the army) use SNA to identify criminal and terrorist networks from traces of communication they collect; and then identify key players in the that se networks 2. Social Network Sites like Facebook use basic elements of SNA to identify and recommend potential friends based on friends-of-friends 3. Civil society organizations use SNA to uncover conflicts of interest in hidden connections between government bodies, lobbies and businesses 4. Network operators (telephony, cable, mobile) use SNA-like methods to optimize the structure and capacity of their networks
- 12. community search protein interaction network https://en.wikipedia.org/wiki/C ommunity_structure Western States Power Grid of USA. Each node represents a community. Node sizes are proportional to the number of nodes in each community and edge sizes are proportional to the amount of edges linking 2 communities. Each node has a different color, to represent a different community. We also use the circular layout, with labels. https://gallery.azure.ai/CustomModule/Graph-Plotting-3
- 13. 1. Whenever you are studying entities relationships, either offline or online, or when you wish to understand how to improve the effectiveness of the network 2. When you want to visualize your data so as to uncover patterns in relationships or interactions 3. When you want to follow the paths that information (or basically anything) and see where they lead to 13
- 15. CNM Social Media Module – Giorgos Cheliotis ([email protected]) 15 How to represent various social networks How to identify strong/weak edges in the network How to identify key/central nodes in network Measures of overall network structure
- 16. CNM Social Media Module – Giorgos Cheliotis ([email protected]) 16 1 2 3 4 1 2 3 4 Graph Anne Jim Mary John Murray Vertex (node) Edge (link) Can we study their interactions as a network? Communication Anne: Jim, tell the Murrays they’re invited Jim: Mary, you and your dad should come for dinner! Jim: Mr. Murray, you should both come for dinner Anne: Mary, did Jim tell you about the dinner? You must come. John: Mary, are you hungry? … Jim Murray
- 17. M Social Media Module – Giorgos Cheliotis ([email protected]) 17 1 2 3 4 1 2 3 4 Graph Anne Jim Mary M John Murray Vertex (node) Edge (link) Communication Anne: Jim, tell the Murrays they’re invited Jim: Mary, you and your dad should come for dinner! Jim: Mr. Murray, you should both come for dinner Mary: Did Jim tell you about the dinner? You must come. John: Mary, are you hungry? … Jim Murray Vertex Vertex 1(Anne) 2 1 3 2(Jim) 3 2 4 3 4 2 1
- 18. 18 1 2 3 4 Graph (directed) Vertex Vertex 1 2 1 3 2 3 2 4 3 4 Edge list Vertex 1 2 3 4 1 - 1 1 0 2 1 - 1 1 3 0 0 - 1 4 0 0 1 - Adjacency matrix
- 19. 19 1 2 3 4 Vertex Vertex 1 2 1 3 2 3 2 4 3 4 Edge list remains the same Vertex 1 2 3 4 1 - 1 1 0 2 1 - 1 1 3 1 1 - 1 4 0 0 1 - Adjacency matrix becomes symmetric 1 2 3 4 Directed Undirected (who knows whom) (who contacts whom) But interpretation is different nowJim
- 20. 20 1 2 3 4 5 6 7 1 2 3 5 4 1 2 54 ‘whole’ network* * no studied network is ‘whole’ in practice; it’s usually a partial picture of one’s real life networks (boundary specification problem) ** ego not needed for analysis as all alters are by definition connected to ego ego alter isolate
- 21. Networks Key Players Cohesion 21 How to represent various social networks How to identify strong/weak ties in the network How to identify key/central nodes in network Measures of overall network structure
- 22. 22 Vertex Vertex Weight 1 2 30 1 3 5 2 3 22 2 4 2 3 4 37 Edge list: add column of weights Vertex 1 2 3 4 1 - 30 5 0 2 30 - 22 2 3 5 22 - 37 4 0 2 37 - Adjacency matrix: add weights instead of 1 Weights could be: • Frequency of interaction in period of observation • Number of items exchanged in period • Individual perceptions of strength of relationship • Costs in communication or exchange, e.g. distance • Combinations of these 1 2 3 4 30 2 37 22 5
- 24. Homophily clusters heterophily strong weak Transitivity cliques Bridges 24 Homophily Strong Weak Transitivity Bridging Interlinked groups Heterophily Cliques Social network TIES CLUSTERING
- 25. Networks Tie Strength Cohesion 25 How to represent various social networks How to identify strong/weak ties in the network How to identify key/central nodes in network Measures of overall network structure
- 26. 26 1 2 3 4 5 6 7 2 3 4 1 4 1 1 Nodes 3 and 5 have the highest degree (4) Values computed with the sna package in the R programming environment. Definitions of centrality measures may vary slightly in other software. Hypothetical graph
- 28. 28 Degree Betweenness Closeness Eigenvector How many people/entity can this person reach directly? How likely is this person to be the most direct route between two people in the network? How fast can this person reach everyone in the network? How well is this person connected to other well- connected people? Centrality measure Interpretation in social networks
- 29. 29 1 2 3 4 5 6 7 0 1.5 6.5 0 9 0 0 Node 5 has higher betweenness centrality than 3 Values computed with the sna package in the R programming environment. Definitions of centrality measures may vary slightly in other software.
- 30. reach 30 1 2 3 4 5 6 7 0.5 0.67 0.75 0.46 0.75 0.46 0.46 Nodes 3 and 5 have the highest (i.e. best) closeness, while node 2 fares almost as well Note: Sometimes closeness is calculated without taking the reciprocal of the mean shortest path length. Then lower values are ‘better’. Values computed with the sna package in the R programming environment. Definitions of centrality measures may vary slightly in other software.
- 31. eigenvector centrality 31 1 2 3 4 5 6 7 0.36 0.49 0.54 0.19 0.49 0.17 0.17 Node 3 has the highest eigenvector centrality, closely followed by 2 and 5 Note: The term ‘eigenvector’ comes from mathematics (matrix algebra), but it is not necessary for understanding how to interpret this measure Values computed with the sna package in the R programming environment. Definitions of centrality measures may vary slightly in other software.
- 32. 32 Degree Betweenness Closeness Eigenvector How many people can this person reach directly? How likely is this person to be the most direct route between two people in the network? How fast can this person reach everyone in the network? How well is this person connected to other well- connected people? Centrality measure Interpretation in social networks
- 33. 33 Degree Betweenness Closeness Eigenvector In network of music collaborations: how many people has this person collaborated with? In network of spies: who is the spy though whom most of the confidential information is likely to flow? In network of sexual relations: how fast will an STD spread from this person to the rest of the network? In network of paper citations: who is the author that is most cited by other well-cited authors? Centrality measure Other possible interpretations…
- 35. Networks Tie Strength Key Players 35 How to represent various social networks How to identify strong/weak ties in the network How to identify key/central nodes in network How to characterize a network’s structure
- 36. EDGEStrength
- 37. 37 1 2 3 4 Reciprocity for network = 0.4
- 38. 38 1 2 3 4 density = #Edges/ (Tot Possible # Edges) 1 2 3 4 density = 5/6 = 0.83 density = 5/12 = 0.42 Edge present in network Possible but not present
- 39. 39 1 2 3 4 5 6 7 1 0.67 0.33 N/a 0.17 N/a N/a clustering coefficient transitivity clique • Clustering algorithms edge betweenness Network clustering coefficient = 0.375 (3 nodes in each triangle x 2 triangles = 6 closed triplets divided by 16 total) Cluster A Cluster B Values computed with the graph package in the R programming environment. Definitions of centrality measures may vary slightly in other software.
- 41. 41 small world local cluster bridge You may have heard of the famous “6 degrees” of separation Sketch of small world structure
- 42. 42 The result is a network with few very highly connected nodes and many nodes with a low degree Such networks are said to exhibit a long-tailed degree distribution And they tend to have a small-world structure! (so, as it turns out, transitivity and strong/weak tie characteristics are not necessary to explain small world structures, but they are common and can also lead to such structures) nodes ordered in descending degree degree short head long tail Example of network with preferential attachment Sketch of long-tailed degree distribution
- 43. 43 We want to be associated with popular people, ideas, items, thus further increasing their popularity, irrespective of any objective, measurable characteristics We evaluate people and everything else based on objective quality criteria, so higher quality nodes will naturally attract more attention, faster Among nodes of similar attributes, those that reach critical mass first will become ‘stars’ with many friends and followers (‘halo effect’) Popularity Quality Mixed model May be impossible to predict who will become a star, even if quality matters Also known as ‘the good get better’ Also known as ‘the rich get richer’
- 44. A useful and relatively simple metric of the degree to which a social network is centralized or decentralized, is the centralization measure In addition to centralization, many large groups and online communities have a core of densely connected users that are critical for connecting a much larger periphery joint degree distributions Bow-tie analysis here CNM Social Media Module – Giorgos Cheliotis ([email protected]) 44 More peripheral clusters and other structures Node s in core
- 45. 45 How can an online social media platform (and its administrators) leverage the methods and insights of social network analysis? How can it encourage a network perspective among its users, such that they are aware of their ‘neighborhood’ and can learn how to work with it and/or expand it? What measures can an online community take to optimize its network structure? Example: cliques can be undesirable because they shun newcomers SNA inspired some of the first SNS’s (e.g. SixDegrees), but still not used so often in conjunction with design decisions – much untapped potential here How can online communities identify and utilize key players for the benefit of the community? What would be desirable structures for different types of online platforms? (not easy to answer)
- 46. 46 • Use the steps outlined in the following pages to visualize and analyze your own network • Think about the key players in your network, the types of ties that you maintain with them, identify any clusters or communities within your network, etc. • Objective: practice SNA with real data! • Present your findings in class next week!
- 47. TouchGraph Google Browser 47 Example TouchGraph Facebook Layout Navigate the graph, examine friend ‘ranks’, friend positions in the network, clusters and what they have in common, try to identify weak and strong ties of yours and assess overall structure of your ego- network https://www.touchgraph.com/navigator
- 48. • Data is more useful when you can extract it from an online platform and analyze with a variety of more powerful tools • Bernie Hogan (Oxford Internet Institute) has developed a Facebook application that extracts a list of all edges in your ego-network (see instructions on next page) • Also, NodeXL (Windows only, see later slide) currently imports data from: Twitter, YouTube, Flickr, and your email client! • Let’s start with installing and learning to use NodeXL… 48
- 49. NodeXL 49 NodeXL sample screenshot Launch Excel and select New -> My Templates ->NodeXLGraph.xltx Go to “Import” and select the appropriate option for the data you wish to import (for Facebook import see next slide first!) NM4881A Tweep Network (weekly data) Click on to ask NodeXL to compute centralities, network density, clustering coefficients, etc. Select to display network graph. You can customize this using and as well as For more info read this NodeXL tutorial
- 50. Launch Excel and open the file that you just saved to your computer. 1. Excel will launch the Text Import Wizard. Select “Delimited”. Click “Next >” 2. Select “Space” as the delimiter, as shown here 3. Select “Next >” and then “Finish”. 4. A new file will be created in Excel. It contains a list of all the nodes in your ego-network, followed by a list of all edges. Scroll down until you find the edges, select all of them and copy them (Ctrl-C) 5. You can now open a new NodeXL file in Excel as explained in the previous slide. Instead of using NodeXL’s import function, paste the list of edges to the NodeXL worksheet, right here 6. In NodeXL select and “Get Vertices from Edge List” 7. Now you can compute graph metrics and visualize your data like explained in the previous slide! 50 This will explain how to export your Facebook data for analysis with a tool like NodeXL here
- 51. CNM Social Media Module – Giorgos Cheliotis ([email protected]) 51
- 52. ilamont ChrisK4u AJC1 52 CNM Social Media Module – Giorgos Cheliotis ([email protected]) Original content in this presentation is licensed under the Creative Commons Singapore Attribution 3.0 license unless stated otherwise (see above)
Editor's Notes
- #4: enterprise value is achieved through enhanced levels of collaboration, more so than simple content consumption. The Enterprise Social Network Graph focusses on people and their relationships with each other. The contrast to the Social Graph is illustrated below: