Structural equation modelling based data fusion for technology forecasting: A generic framework

Abstract

Technology Intelligence (TI) involves the process of capturing technology related data, converting this data into information by determining relational connections and refining information to produce knowledge that can guide strategic decision makers. Technology indicators are those sources of technology related data that allow for the direct characterisation and evaluation of technologies over their whole life cycle. Future-oriented Technology Analysis (FTA), which is a forward-looking approach in scrutinizing the information that has been distilled from a set of technology indicators, can potentially provide decision makers with useful Technology Forecasting (TF) knowledge. The paper postulates that TF can be viewed as an instance of Data Fusion (DF), which is a formal framework that defines tool, as well as the application of these tools, for the unification of data originating from different sources. Within the field of DF relational connections define context. Context sensitive DF techniques refine the generated knowledge based on the characteristics of exogenous context related variables. Structural Equation Modelling (SEM), which is a statistical technique capable of evaluating complex hierarchical dependencies between latent and observed problem and context variables, has been shown to be effective in implementing context sensitive DF. In the paper a generic framework is introduced for SEM based DF of technology indicators in order to produce TF output metrics. The paper also provides the research methodology that will be used in a future study to evaluate the validity of the generic framework for the case of National Research and Education Networks (NRENs).