Lines 1-27
Link Here
|
1 |
From the README: |
1 |
This package consists of Perl modules along with supporting Perl programs |
2 |
|
2 |
that implement the semantic relatedness measures described by Leacock |
3 |
This package consists of Perl modules along with supporting Perl programs that |
3 |
Chodorow (1998), Jiang Conrath (1997), Resnik (1995), Lin (1998), Hirst St |
4 |
implement the semantic relatedness measures described by Leacock Chodorow |
4 |
Onge (1998), Wu Palmer (1994), the adapted gloss overlap measure by |
5 |
(1998), Jiang Conrath (1997), Resnik (1995), Lin (1998), Hirst St Onge (1998) |
5 |
Banerjee and Pedersen (2002), and a measure based on context vectors |
6 |
and the adapted gloss overlap measure by Banerjee and Pedersen (2002). The Perl |
6 |
by Patwardhan (2003). The details of the Vector measure are described in the |
7 |
modules are designed as object classes with methods that take as input two word |
7 |
Master's thesis work done by Patwardhan (2003) at the University of Minnesota |
8 |
senses. The semantic relatedness of these word senses is returned by these |
8 |
Duluth. The Perl modules are designed as objects with methods that take as |
9 |
methods. A quantitative measure of the degree to which two word senses are |
9 |
input two word senses. The semantic relatedness of these word senses is |
10 |
related has wide ranging applications in numerous areas, such as word sense |
10 |
returned by these methods. A quantitative measure of the degree to which two |
11 |
disambiguation, information retrieval, etc. For example, in order to determine |
11 |
word senses are related has wide ranging applications in numerous areas, such |
12 |
which sense of a given word is being used in a particular context, the sense |
12 |
as word sense disambiguation, information retrieval, etc. For example, in |
13 |
having the highest relatedness with its context word senses is most likely to |
13 |
order to determine which sense of a given word is being used in a particular |
14 |
be the sense being used. Similarly, in information retrieval, retrieving |
14 |
context, the sense having the highest relatedness with its context word |
15 |
documents containing highly related concepts are more likely to have higher |
15 |
senses is most likely to be the sense being used. Similarly, in information |
16 |
precision and recall values. |
16 |
retrieval, retrieving documents containing highly related concepts are more |
|
|
17 |
likely to have higher precision and recall values. |
17 |
|
18 |
|
18 |
A command line interface to these modules is also present in the package. The |
19 |
A command line interface to these modules is also present in the package. The |
19 |
simple, user-friendly interface returns the relatedness measure of two given |
20 |
simple, user-friendly interface returns the relatedness measure of two given |
20 |
words. A number of switches and options have been provided to modify the output |
21 |
words. |
21 |
and enhance it with trace information and other useful output. Details of the |
|
|
22 |
usage are provided in other sections of this README. Supporting utilities for |
23 |
generating information content files from various corpora are also available in |
24 |
the package. The information content files are required by three of the |
25 |
measures for computing the relatedness of concepts. |
26 |
|
22 |
|
27 |
WWW: http://search.cpan.org/dist/WordNet-Similarity/ |
23 |
WWW: http://search.cpan.org/dist/WordNet-Similarity/ |