Difference between revisions of "Amis/Disambiguation"
| (13 intermediate revisions by 2 users not shown) | |||
| Line 1: | Line 1: | ||
| + | ling073-ami github: https://github.swarthmore.edu/Ling073-sp23/ling073-ami | ||
| + | |||
==initial evaluation of ambiguity== | ==initial evaluation of ambiguity== | ||
| − | to remove weird characters from tests, run with options like this: <br> | + | to remove weird characters from tests, run command line measurements with options like this: <br> |
cat ami.corpus.basic.txt | apertium-destxt | hfst-proc ./ami.automorf.hfst | apertium-retxt | sed 's/$\W*\^/$\n^/g' | wc -l | cat ami.corpus.basic.txt | apertium-destxt | hfst-proc ./ami.automorf.hfst | apertium-retxt | sed 's/$\W*\^/$\n^/g' | wc -l | ||
| + | |||
| + | ex., to get all of the ambiguous analyses do: <br> | ||
| + | cat ami.corpus.basic.txt | apertium-destxt | hfst-proc ./ami.automorf.hfst | apertium-retxt | sed 's/$\W*\^/$\n^/g' | grep '\/.*\/' | ||
The ambiguity of a corpus is the average number of analyses provided per token. Our initial ambiguity is 2066/1941 = 1.064 | The ambiguity of a corpus is the average number of analyses provided per token. Our initial ambiguity is 2066/1941 = 1.064 | ||
| Line 27: | Line 32: | ||
| + | ==Sentences== | ||
| + | For genitive vs. possessive disambiguation. | ||
| + | Possessive only appears directly after noun marker ({{tag|nm}}). | ||
| + | |||
| + | * Ulahen namu ku ina namu. (genitive) https://shs.hal.science/halshs-03483275 (p. 39) | ||
| + | * Ngaʼay ho ku namu? (Posessive) Wu Dissertation p.86. | ||
| + | |||
| + | echo "Ulahen namu ku ina namu" | apertium -d. ami-morph <br> | ||
| + | ^Ulahen/ulah<v><uv><futi>$ ^namu/'''amu<prn><pers><pl><p2><gen>'''/amu<prn><pers><pl><p2><pos>$ ^ku/u<nm><un><nom>$ ^ina/ina<n>$ ^namu/'''amu<prn><pers><pl><p2><gen>'''/amu<prn><pers><pl><p2><pos>$ | ||
| + | |||
| + | echo "Ngaʼay ho ku namu" | apertium -d. ami-morph <br> | ||
| + | ^Ngaʼay/ngaʼay<adj>$ ^ho/ho<asp><impf>$ ^ku/u<nm><un><nom>$ ^namu/'''amu<prn><pers><pl><p2><pos>'''/amu<prn><pers><pl><p2><gen>$ | ||
| + | ==final evaluation of ambiguity== | ||
| + | The number of tokens is 2114. The total number of analyses is 2239. The ambiguity is 2239/2114 = 1.059 | ||
[[Category:Amis]][[Category:Sp23_Disambiguation]] | [[Category:Amis]][[Category:Sp23_Disambiguation]] | ||
Latest revision as of 13:58, 30 March 2023
ling073-ami github: https://github.swarthmore.edu/Ling073-sp23/ling073-ami
initial evaluation of ambiguity
to remove weird characters from tests, run command line measurements with options like this:
cat ami.corpus.basic.txt | apertium-destxt | hfst-proc ./ami.automorf.hfst | apertium-retxt | sed 's/$\W*\^/$\n^/g' | wc -l
ex., to get all of the ambiguous analyses do:
cat ami.corpus.basic.txt | apertium-destxt | hfst-proc ./ami.automorf.hfst | apertium-retxt | sed 's/$\W*\^/$\n^/g' | grep '\/.*\/'
The ambiguity of a corpus is the average number of analyses provided per token. Our initial ambiguity is 2066/1941 = 1.064
The number of all forms is 6283. The number of unique forms is 6273.
The comparison:
40,41d39
< ako
< aku
68d65
< iso
70,71d66
< isu
< ita
4827,4828d4821
< nami
< namo
4831,4832d4823
< namu
< nangra
5020d5010
< ningra
Sentences
For genitive vs. possessive disambiguation. Possessive only appears directly after noun marker (<nm>).
- Ulahen namu ku ina namu. (genitive) https://shs.hal.science/halshs-03483275 (p. 39)
- Ngaʼay ho ku namu? (Posessive) Wu Dissertation p.86.
echo "Ulahen namu ku ina namu" | apertium -d. ami-morph
^Ulahen/ulah<v><uv><futi>$ ^namu/amu<prn><pers><pl><p2><gen>/amu<prn><pers><pl><p2><pos>$ ^ku/u<nm><un><nom>$ ^ina/ina<n>$ ^namu/amu<prn><pers><pl><p2><gen>/amu<prn><pers><pl><p2><pos>$
echo "Ngaʼay ho ku namu" | apertium -d. ami-morph
^Ngaʼay/ngaʼay<adj>$ ^ho/ho<asp><impf>$ ^ku/u<nm><un><nom>$ ^namu/amu<prn><pers><pl><p2><pos>/amu<prn><pers><pl><p2><gen>$
final evaluation of ambiguity
The number of tokens is 2114. The total number of analyses is 2239. The ambiguity is 2239/2114 = 1.059
