You are here

John Williams

DBA Graduate - 2008

Thesis title

discerning industrial activity in NE England - networks and cluster through the lens of the internet


Dimitris Assimakopoulos
Research on industrial activity has taken many forms, as both practitioners and academics have sought to gain an understanding of the drivers of the economy in any particular sphere of interest. One facet of this research effort attempts to determine industrial clusters or particular groupings of economically linked firms. Although the tools available to researchers have improved, they still rely heavily on data gathered from databases that describe the firm’s activity. These data sources suffer significant shortcomings, especially if they rely on standard industry classification (SIC) codes. However, the rise of the Internet and its vast capacity as an information source has created the possibility of gaining new insights into the activity of firms and the way they interact. This thesis therefore investigates the usefulness of the Internet and World Wide Web as information sources for the study of industry in general and clusters and networks in particular. It begins with an analysis of the practicality of using a corpus of regional company URLs to extract descriptive text from each firm’s Web site. The resultant database of keywords and descriptors of firms’ activity, markets, and connections can be investigate to determine industrial activity across a region. This insight is much richer in detail and depth than that obtained from coarse-grained SIC descriptors. Next, this study uses the definitions associated with industrial clusters to investigate the connections between firms and other non-traded dependencies. Publicly available data sources on the Internet reveal embedded links, visible connections to external references, and web-derived in-links (i.e., third parties that reference the firm on their own Web sites). Through this effort, the author can draw graphic connecting networks and derive the interesting conclusion that, at least in the North East of England, the clustering pertains not so much to artifacts per se but rather to competencies in a wide range of sectors that share both common antecedents and current practices, such as strong engineering skills for the design and manufacture of large structures that operate in difficult or even hostile environments. Overall then, the Internet offers additional insights compared with more conventional forms of determining industrial clusters. These insights should be as complements, rather than substitute, methods. The tools available for Internet searches are continually evolving, as is the provision and use of corporate Websites by all sectors of industry and firm sizes. A logical conclusion of this process implies that in the not-too-distant future, the prospects for the gleaning of information from Web-based processes will be significantly enhanced.