Japhug/Disambiguation

Initial Evaluation of Ambiguity

Level of ambiguity in our corpus: ~1.09 (1.09 analyses per form)

An ambiguous term in our corpus: ndʑiwa/wa<n><px2du>/wa<n><px3du (This is not ideal, since the difference isn't huge)

Two sentences with possible readings of an ambiguous term:

• nɤʑo sɯku ɕtɤkre ɕpɯkre tɕɤn aʑo kɯ tunɯrdoʁa

will be analyzed as

"<nɤʑo>"
        "nɤʑo" prn
;       "ʑo" v iv trop REMOVE:26
"<sɯku>"
        "sɯku" n ind
"<ɕpɯkre>"
        "*ɕpɯkre"
"<tɕɤn>"
        "*tɕɤn"
"<aʑo>"
        "aʑo" prn
        "ʑo" v iv pass
"<kɯ>"
        "kɯ" post
"<tunɯrdoʁa>"
        "*tunɯrdoʁa"
"<.>"
        "." sent

in which word nɤʑo can be either a pronoun of a tropative verb. Here it must be a pronoun.

• ..

another possible candidate: the verb tu. It can mean "be", "exist" which is semantically similar, or "be in the presence of" in specific cases (if in the infinitive form kɯ-tu, directly following a noun or pronoun).

Console

Sort these out later

eevee[ling073-jya]$ cat jya.corpus.basic.txt | apertium-destxt | hfst-proc jya.automorf.hfst | apertium-retxt | sed 's/$\W*\^/$\n^/g' | grep '\/.*\/' | sort | uniq -c | sort -n

     1 #^ja/ja<v><iv>/ja<v><tv>$
     1 #^tɕhi/tɕhi<n><ind>/tɕhi<prn>$
     1 ^alo/alo<adv>/lo<v><iv><pass>$
     1 ^aʑɯɣ/ʑɯɣ<v><iv><pass>/ʑɯɣ<v><tv><pass>$
     1 ^kho/kho<n><ind>/kho<v><tv>$
     1 ^kɤsɯso/so<v><iv><caus><prc><obj><ind>/sɯso<v><tv><prc><obj><ind>$
     1 ^kɤznɯna/nɯna<v><iv><caus><prc><obj><ind>/znɯna<v><tv><prc><obj><ind>$
     1 ^mu/mu<adv>/mu<v><iv>$
     1 ^mɤʑɯ/mɤʑɯ<adv>/ʑɯ<v><iv><aor><neg>$
     1 ^mɯkɤnɯʑɯβ/nɯʑɯβ<v><iv><prc><obj><px2pl>/nɯʑɯβ<v><iv><prc><obj><px3pl>$
     1 ^ndʐa/ndʐa<adv>/ndʐa<n><ind>$
     1 ^ndʑikɤcha/cha<v><iv><prc><obj><px2du>/cha<v><iv><prc><obj><px3du>$
     1 ^ndʑikɤkho/kho<v><tv><prc><obj><px2du>/kho<v><tv><prc><obj><px3du>$
     1 ^ndʑipɕi/pɕi<n><px2du>/pɕi<n><px3du>$
     1 ^ndʑiwa/wa<n><px2du>/wa<n><px3du>$
     1 ^nɤjaʁ/jaʁ<v><iv><trop>/nɤjaʁ<v><tv>$
     1 ^nɤmbro/mbro<n><px2sg>/mbro<v><iv><trop>$
     1 ^pjɤsɯso/so<v><iv><caus><pii>/sɯso<v><tv><pii>$
     1 ^qartsɯ/qartsɯ<clf>/qartsɯ<n><ind>$
     1 ^sɯpa/pa<v><iv><caus>/pa<v><tv><caus>/sɯpa<n><ind>/sɯpa<v><tv>$
     1 ^ɯkɤndza/kɤndza<n><px3sg>/ndza<v><tv><prc><obj><px3sg>$
     1 ^ɯkɯznɤja/nɤja<v><iv><caus><prc><sbj><px3sg>/znɤja<v><tv><prc><sbj><px3sg>$
     2 ^///<sent>$
     2 ^ambro/mbro<n><px1sg>/mbro<v><iv><pass>$
     2 ^asci/sci<n><px1sg>/sci<v><iv><pass>$
     2 ^kɤndza/kɤndza<n><ind>/ndza<v><tv><prc><obj><ind>$
     2 ^nɤkɯɕaβ/ɕaβ<v><iv><prc><sbj><px2sg>/ɕaβ<v><tv><prc><sbj><px2sg>$
     2 ^smi/smi<n><ind>/smi<v><iv>$
     2 ^sŋi/sŋi<clf>/sŋi<n><ind>$
     3 ^ata/ata<v><iv>/ta<n><px1sg>/ta<v><tv><pass>$
     3 ^fse/fse<v><iv>/fse<v><tv>$
     3 ^qaʁ/qaʁ<n><ind>/qaʁ<v><tv>$
     4 ^maʁ/maʁ<cnj>/maʁ<n><ind>$
     4 ^ŋgɯ/ŋgɯ<n><ind>/ŋgɯ<v><iv>$
     5 ^ndʑithɤβ/thɤβ<n><px2du>/thɤβ<n><px3du>$
     6 ^me/me<post>/me<v><iv>$
     6 ^stu/stu<v><iv>/stu<v><tv>$
     9 ^kha/kha<clf>/kha<n><ind>$
    10 ^li/li<adv>/li<n><ind>$
    10 ^tɕhi/tɕhi<n><ind>/tɕhi<prn>$
    12 ^kɯfse/fse<v><iv><prc><sbj><ind>/fse<v><tv><prc><sbj><ind>$
    13 ^nɤʑo/nɤʑo<prn>/ʑo<v><iv><trop>$
    16 ^aʑo/aʑo<prn>/ʑo<v><iv><pass>$
    17 ^mbro/mbro<n><ind>/mbro<v><iv>$
    32 ^ra/ra<cnj>/ra<v><iv>$
    43 ^ci/ci<other>/ci<v><tv>$
    43 ^ma/ma<other>/ma<post>$
    61 ^ndɤre/ndɤre<cnj>/ndɤre<other>$
    67 ^ʑo/ʑo<other>/ʑo<v><iv>$
    78 ^kɯ/kɯ<other>/kɯ<post>$
   182 ^nɯ/nɯ<adv>/nɯ<other>$