ISR11
Scientific Report No. ISR-11 Information Storage and Retrieval
Design Criteria for Automatic Information Systems
chapter
M. E. Lesk
G. Salton
Harvard University
Gerard Salton
Use, reproduction, or publication, in whole or in part, is permitted for any purpose of the United States Government.
v-b
with the SMART system:
a) IRE - 1 : a set of about 4oo abstracts of documents in the
computer literature published in 1959, used with approximately
20 search requests;
b) IRE - 2 a set of about 400 abstracts of documents in the
computer literature published in 1960 and 1961, used with
approximately 20 search requests;
c) ADI : a set of 82 short papers in documentation, each
approximately 2000 words long, presented at the 1963 Annual
Meeting of the American Documentation Institute, and processed
against 35 search requests;
d) Cranfield - 1 : a set of 200 abstracts of documents in aero-
nautical engineering previously used by the Aslib-Cranfield
project L12], and processed against [OCRerr]2 search requests;
e) Cranfield - 2 : a set of 1200 additional document abstracts
in aeronautical engineering, similar to the abstracts included
in the preceding collection.
It is seen that these collections fall into three distinct subject
areas: computer science, documentation, and aeronautical engineering.
The ADI collection in documentation is of particular interest because full
papers are available rather than only document abstracts. The Cranfield
collections, on the other hand, are the only ones which are also manually
indexed by subject experts, thus permitting a comparison of the standard
keyword search procedures with the automatic text processing methods.
The evaluation results obtained with the first four of these collec-
tions are summarized in the next section.