ABSTRACT
As mentioned in Chapter 7, one of the major tasks in social network analysis is to represent the network. Typically, social networks are represented using graphs. One needs to come up with a way to electronically represent these graphs. Another task is to build the network. For example, in social networks such as Facebook, one can build a graph where the nodes are the people in the network and the links are the relationships between the people (e.g., friendships). In many situations, we may need to build a network from vast amounts of data, both structured and unstructured. The data could emanate from various databases as well as e-mails, blogs, and web pages. Therefore, we need to use various analytics techniques to extract the nuggets from the various data sources and then build the network. Once the network is developed, we can carry out analysis to extract information such as communities of interest and those who are leaders of the network. Recently, semantic web technologies are being applied to represent the graph structures presenting social networks (Mika, 2007). One of the prominent ontologies that have been developed based on Resource Description Framework (RDF) is FOAF (friend of a friend), which has been used to build a very large social network. We need techniques to extract graph structures from multimodal data, integrate the graph structures, and mine or analyze the structures and extract patterns.
