Industrial Clusters in England

Methods
This study uses a novel data-driven approach to investigate the patterns of geographical clustering and functional integration across three sectors: Digital-Health, Financial Services and Processing Industry. Machine learning techniques are used to collect unstructured information from the websites of companies that belong to a particular sub-sector or group. Based on this information companies are then classified into the three sectors under investigation. Density based clustering algorithms are used to identify the patterns of geographical agglomeration of these companies. The analysis departs considerably from previous works on UK clusters. First, by using Internet data to identify which companies belong to each sector the approach is mostly data-driven and is less reliant on SIC codes. This allows us to classify as part of a sector companies with different SIC codes. Second, by using an algorithm that identifies clusters based on the physical distance between companies we can identify clusters spreading over multiple discrete administrative areas. Third, by investigating the links included on companies’ websites it becomes possible to investigate the relationships between companies and between companies and other institutions.
Timescale and funder
This project is being carried out over the period December 2015 to April 2016 and is funded by the Department for Business, Innovation and Skills.
Project team
This project is being carried out by NIESR in collaboration with SpazioDati and City REDI.
Infographic by Strelnik: