Understanding edge lists for relational data storage

Example one-mode network

Traditionally, relational data is stored in an adjacency matrix.  An adjacency matrix is a way of representing relational data by the presence or absence of a tie. This tie is listed as either a 0, which represents an absence of a tie, or a 1, which represents the presence of a tie. Ties can also [...]

Lab 6: Dynamic Network Analysis

Meta-network

  The dynamic analysis of networks takes many forms. In this lab we will examine how it is implemented in ORA (Organizational Risk Analyzer). At the heart of ORA’s approach is the idea of a meta-matrix of networks that not only includes social networks but knowledge networks (who knows what), information networks (what ideas are [...]

Lab 5: Fusing Geospatial and Relational Data

geospatialfuse

In this lab we will examine how ORA fuses geospatial and relational data so that they can be analyzed concurrently. This is a relatively recent development within social network analysis, and it appears to hold great promise. Other efforts are focused on including both social network and geospatial data into single multivariate regression statistical analyses. We won’t consider [...]

Lab 4: Brokers and Bridges

brokers_bridges

Bridges and brokers can be thought of as elements of a network that connect or hold together the subnetworks of the larger network. Brokers are the actors that connect the subnetworks together, while bridges are the ties between the brokers who hold the network together. Thus, actors who are considered to be brokers are similar [...]

Lab 3: Identifying Subgroups

clustering

A major focus of social network analysis is to identify dense clusters of actors “among whom there are relatively strong, direct, intense, and/or positive ties.”1 Social network analysts often refer to these clusters of actors as cohesive subgroups and generally assume that “social interaction is the basis for solidarity, shared norms, identity, and collective behavior, so [...]

Lab 2: Centrality and Power

centrality

Centrality is one of the oldest concepts in social network analysis and is used quite extensively (and often uncritically) in analyses of dark networks. A central actor can be seen as someone who has a lot of ties to other actors (degree centrality), as someone who is closer (in terms of path distance) to all [...]

Lab 1: Prepping and Visualizing Networks

network_visualization

Network visualizations can help us trace and detect patterns of relations in a given social network. However, the human eye is easily fooled, which is why social network analysts rely on graph drawing algorithms that systematically allocate vertices in “social space.” One of the purposes of  this exercise is to introduce you to the capabilities [...]

ORA walkthrough

ORA main window

  The Organizational Risk Analyzer (ORA) is a powerful social and dynamic network analysis tool from the Carnegie Mellon CASOS lab.  It is one of several applications we use in the CORE lab to build our analytic products and teach students. You can download the latest version of ORA for free (for non-commercial use) at [...]

Social network theory presentations

Social Network Analysis

These presentations were given during the last CORE lab training session. These three presentations cover some of the theory that motivates the work in social network analysis and provide an introduction to the terminology that will be used repeatedly in the labs. You can download a PDF of each of the presentations in addition to [...]