SSRN Home Search and Download Papers Browse Abstract and Paper Submission Subscribe to Networks View Briefcase Top Papers Top Authors Top Institutions

 

Abstract

 
 

Footnotes (83)

Beta

 


 



Does Search Advisor Depend Too Little on Classifiers, and Too Much on Algorithms, for Headnote Classification? Evidence on the Perils of Search Advisor's Automation and the Virtues of Thomson-West Classifiers

Michael Ginsborg
Law Librarian


August 1, 2007


Abstract:     
Search Advisor has just been renamed "Search by Topic or Headnote" to accommodate topical research of not just case law, but other sources of primary and secondary law that users may combine. I retain the name "Search Advisor" to signify the initial application of LexisNexis classification to a digest for case law - an application now represented by "Search by Headnote." In this three-part series, I present evidence that Search Advisor largely depends on word-matching-and-ranking algorithms for headnote classification. If my conjecture is right, then Search Advisor has a fatal flaw in design that compromises its usefulness as a classification system. By substituting algorithms for much of the work that we should expect classifiers to do, Search Advisor collapses from the weight of its automation. Whether my conjecture is right or wrong, I hope to renew appreciation of how classifiers make the Key Number System a uniquely valuable tool of case law research.

Keywords: Headnotes, Digest, Key Number, Search Advisor, Westlaw, LexisNexis

JEL Classifications: K00, K10, K40

Working Paper Series

Date posted: February 19, 2009 ; Last revised: February 19, 2009

Suggested Citation

Ginsborg, Michael, Does Search Advisor Depend Too Little on Classifiers, and Too Much on Algorithms, for Headnote Classification? Evidence on the Perils of Search Advisor's Automation and the Virtues of Thomson-West Classifiers (August 1, 2007). Available at SSRN: http://ssrn.com/abstract=1345465


Export to: Export Citation What's this?

Contact Information

Michael Ginsborg (Contact Author)
Law Librarian ( email )
Alameda, CA
Feedback to SSRN (Beta)


Paper statistics
Abstract Views: 117
Downloads: 34
Footnotes: 83

© 2010 Social Science Electronic Publishing, Inc. All Rights Reserved.  FAQ   Terms of Use   Privacy Policy   Copyright
This page was served by apollo 6 in 0.141 seconds.