Unpacking Big Systems -- Natural Language Processing Meets Network Analysis. A Study of Smart Grid Development in Denmark.

Abstract

Studies within the detection of technological trajectories and technology forecasting tend traditionally to rely on patent or bibliometric data. The main drawback of these invention-focused approaches is their inability to account for many mainly non-technical factors related to the social and institutional framing of technology. Value driven policies, technological and institutional path dependencies or user expectations and routines have major impact on the technological outcomes in a particular context. This paper suggests a new method for the mapping and analysis of large (technical) systems and contained technological trajectories on a national level using a combination of methods from statistical natural language processing, vector space modelling and network analysis. The proposed approach does not aim at replacing the researcher or expert but rather offers the possibility to algorithmically structure and to some extent quantify unstructured text data. The utilized filtered corpora consist of two types of Danish text-documents: 99 R&DD project descriptions and 574 (initially before filtering 813) non-academic/industrial journal publications dealing with the development of the smart energy grid in Denmark. Results show that in the explored case it is not mainly new technologies and applications that are driving change but innovative re-combinations of old and new technologies.