Difference between revisions of "Fijian/Transducer"
From LING073
(→Notes) |
(→Analyser Evaluation) |
||
| Line 11: | Line 11: | ||
:-Coverage: 4.67% | :-Coverage: 4.67% | ||
| − | * | + | *Initial top unknown words: |
na 113 | na 113 | ||
| Line 55: | Line 55: | ||
**The top unknown words ''na'', ''a'', and ''o'' are articles {{tag|art}}: ''na'' and ''a'' are used with common nouns and ''o'' is used with pronouns and proper names. | **The top unknown words ''na'', ''a'', and ''o'' are articles {{tag|art}}: ''na'' and ''a'' are used with common nouns and ''o'' is used with pronouns and proper names. | ||
**The unknown words ''me'' is a relator {{tag|rel}}. | **The unknown words ''me'' is a relator {{tag|rel}}. | ||
| + | |||
| + | *After adding the analyses of ''na'', ''a'', ''o'', and ''me'' to my transducer, the coverage went from 4.67% to 24.35%. | ||
==Generator Evaluation== | ==Generator Evaluation== | ||
Revision as of 14:49, 1 March 2018
Contents
Transducer Link
The morphological transducer for Fijian is at: https://github.swarthmore.edu/hwang11/ling073-fij
Analyser Evaluation
- Total number of stems in transducer: 21
- Current Coverage: Running aq-covtest on my fij.corpus.basic.txt, I got:
- -Number of tokenised words in the corpus: 1306
- -Coverage: 4.67%
- Initial top unknown words:
na 113
a 75
e 62
o 40
ni 40
mai 35
ka 33
i 30
sa 28
E 23
ko 19
vei 18
me 17
u 16
dua 13
kina 11
ki 10
Viti 10
koya 9
vaka 9
- The top unknown words na, a, and o are articles <art>: na and a are used with common nouns and o is used with pronouns and proper names.
- The unknown words me is a relator <rel>.
- After adding the analyses of na, a, o, and me to my transducer, the coverage went from 4.67% to 24.35%.
Generator Evaluation
Initial evaluation of morphological generation
Notes
When running "morph-test" with the code "morph-test -csi fij-yaml | most", I always get an error message saying that "the program 'most' is currently not installed".
