<DOC>
<DOCNO> IRS13 </DOCNO>
<TITLE> Scientific Report No. IRS-13 Information Storage and Retrieval </TITLE>
<SUBTITLE> Thesaurus, Phrase and Hierarchy Dictionaries </SUBTITLE>
<TYPE> chapter </TYPE>
<PAGE CHAPTER="7" NUMBER="26">
<AUTHOR1> E. M. Keen </AUTHOR1>
<PUBLISHER> Harvard University </PUBLISHER>
<EDITOR1> Gerard Salton </EDITOR1>
<COPYRIGHT MTH="December" DAY="" YEAR="1967" BY="National Science Foundation">
Use, reproduction, or publication, in whole or in part, is permitted for any purpose of the United States Government.
</COPYRIGHT>
<BODY>
Stem(s) versus Thesaurus(t)
Input, Matching Function Precision at Recall Normalized
No. Collection and Thesaurus Type .2 .5 .8 Rec. Pre.
20 Title, Cosine Numeric, Thes-l []
21 Title, Cosine Numeric, Thes-SAl S T T T T
22 ADI Text, Cosine Logical, Thes-l S []S S T S
23 35 Text, Cosine Logical, Thes-SAl [[OCRerr] S
24 Requests Abstract, Cosine Logical, Thes-l T T T T
25 (contd) Text, Overlap Logical, Thes-l fls S []S
26 Text, Overlap Logical, Thes-SAl [[OCRerr]]
27 Abstract, Overlap Logical, Thes-l T S T T T
The difference of one dictionary merit over the other is [OCRerr]ller than 0.05 when a letter
alone appears, and larger than 0.05 when letters are circled.
Table Summarizing 18 Precision Versus Recall Plots not Presented, Comparing
Stem and Thesaurus Dictionaries on Three Collections
Fig. 17
(continued)
</BODY>
</PAGE>
</DOC>