Financial distress of U.S. Banking industry viewed through web data

Authors

E. Romero-Frías, Vaughan, L.; Faculty of Information and Media Studies, University of Western Ontario, London, ON N6A 5B7, Canada; email: lvaughan@uwo.caFollow
L. Vaughan

Document Type

Conference Proceeding

Publication Date

1-1-2009

Journal

12th International Conference on Scientometrics and Informetrics, ISSI 2009

First Page

206

URL with Digital Object Identifier

https://doi.org/

Last Page

210

Abstract

Building on previous studies on inlink and co-link research on commercial websites, the current study attempted to apply and combine both methods to investigate the recent financial crisis in the U.S. banking industry. The methods combine Web content mining and Web structure mining. Two sets of inlink and co-link data were collected, one with the keywords "crisis", "bailout" and "subprime" and one without any keyword. The one with the keywords was meant to include only inlink webpages that are likely to be about the financial crisis. The number of inlink pages that contained the keywords correlates significantly with the degree of a bank's crisis measured by the amount of government bailout money. Both sets of co-link data were analysed using multidimensional scaling to generate maps of business competition. The comparison between the maps with and without keywords shows that the map generated from the co-link data with keywords depicts a more accurate image of the industry by clustering banks with more financial problems together.