About | HeinOnline Law Journal Library | HeinOnline Law Journal Library | HeinOnline

1 J.L. & Inf. Sci. 131 (1981)
The Use of Citation Vectors for Legal Information Retrieval

handle is hein.journals/jlinfos1 and id is 147 raw text is: Vol. 1 No. 2

THE USE OF CITATION VECTORS FOR LEGAL
INFORMATION RETRIEVAL
C. TAPPER
Colin Tapper is one of the founders of the study of
computers and law in England, and this paper adds to his
contributions to the field. As its title indicates, the article
is concerned with the use of case citations as selection vectors
in legal information retrieval, and, in particular, with the value
of citation vestors in comparison to the usual semantic vectors
currently used.
The author details recent experiments with citation vectors
in the United States and at the Norwegian Research Centre
for Computers and Law (NRCCL). The comparative results
of using citation vectors against semantic vectors in these
experiments are documented and considered, and Mr. Tapper
provides some valuable discussion of the algorithms used in
computing and assessing vectors in data retrieval. Despite the
complexity of this work, it will be of great value to all
interested in the field of computers and law, because of its
implications for the future development of legal data retrieval.
The first section of this article is intended for those who have no,
or little, previous awareness of legal information retrieval techniques.
Since the main aim of the article is to explain the theory behind the
substitution for such methods of citation vectors those who have the
requisite familiarity with matching and vector based systems as applied
to law might prefer to start with the second section.
1. Current Legal Information Retrieval Techniques
It is now about 25 years since at the University of Pittsburgh
in Pennsylvania Professor John Horty first succeeded in applying
computerised methods to the retrieval of legal information. It is a
tribute to his insight that the techniques which he devised remain the
bedrock of virtually all of the systems which operate in the world
today. The essence of the technique is the identification in the text
of a document of a word, or words, in a particular combination which
have been selected by the lawyer as being likely to indicate the
relevance of that document to the lawyer's problem. As normally
implemented the system creates a concordance of the full legal texts
constituting the database of the system, excluding only words of such
low prima facie information content that they are highly unlikely to
be nominated by lawyers as search terms. Each concordance item
then becomes a potential search term, and searches are typically
conducted by the nomination of classes of words, for example synonyms,
grammatical variations, particularisations and generalisations, which
must occur in a given relationship to other similar classes in a docu-
ment in order for it to satisfy the search request as a potentially

What Is HeinOnline?

HeinOnline is a subscription-based resource containing nearly 3,000 academic and legal journals from inception; complete coverage of government documents such as U.S. Statutes at Large, U.S. Code, Federal Register, Code of Federal Regulations, U.S. Reports, and much more. Documents are image-based, fully searchable PDFs with the authority of print combined with the accessibility of a user-friendly and powerful database. For more information, request a quote or trial for your organization below.



Short-term subscription options include 24 hours, 48 hours, or 1 week to HeinOnline with pricing starting as low as $29.95

Contact us for annual subscription options:

Already a HeinOnline Subscriber?

profiles profiles most