Fundamentals of Big Data Network Analysis for Research and Industry

4Research Methods Using SNA

Research related to network analysis starts from recognizing research problems through empirical studies and developing hypotheses to prove empirical evidence. To set up a hypothesis, attribute variables as well as related variables among actors can be considered. In performing a network analysis, not only are related variables considered, but it is also possible to set up a hypothesis considering a variety of aspects including relational features among actors, which can affect the actors’ attribute variable and have an effect on the actors’ relations. A research plan is constructed by considering the network modeling of defining nodes and links within the network as well as determining a means of collecting data. Depending on the form of network data that would be collected by a researcher, a variety of methods to collect data, including a survey, interview, observation, web-log, bibliographic data, and so on, can be used. Collected data are basically expressed within the matrix, and data cleansing is necessary to make it available for network analysis. The process of data cleansing is essential, as unstructured data have recently been gathered and used to a greater extent. In addition, there is a process to confirm whether a hypothetical relationship is supported, which is formed by using various indexes and techniques of network analysis. Finally, social network analysis can be completed after interpretation of the research results to draw a ...