Belarusian and Russian/Final report - Revision history

Kvld: /* Future work */

2016-08-22T21:42:26Z

Future work

2016-08-22T21:42:11Z

Future work

2016-08-22T21:27:30Z

Statistics

2016-08-22T21:21:54Z

Testvoc

2016-08-22T21:21:40Z

Evaluation

2016-08-22T20:59:53Z

Statistics

2016-08-22T20:58:57Z

Statistics

2016-08-22T20:23:27Z

Russian

2016-08-22T20:13:32Z

Belarusian

2016-08-22T09:37:53Z

Post-generation

@@ Line 95: / Line 95: @@
 For the future I'm going to continue working on this project. Main goals:
 # Finish working on bilingual dictionary and monolingual dictionaries, get clean testvoc for bel→rus.
-# Fix existsing errors in Russian dictionary.
+# Fix existing errors in Russian dictionary.
 # Add more transfer rules.
 # Add missing forms for adjectives and participles to Belarusian dictionary.

@@ Line 93: / Line 93: @@
 =Future work=
+For the future I'm going to continue working on this project. Main goals:
+# Finish working on bilingual dictionary and monolingual dictionaries, get clean testvoc for bel→rus.
+# Fix existsing errors in Russian dictionary.
+# Add more transfer rules.
+# Add missing forms for adjectives and participles to Belarusian dictionary.
 =References=

@@ Line 28: / Line 28: @@
 =Statistics=
 ===Commits===
+Full list of my contribution (commits, diffs) extracted from SVN: https://apertium.projectjj.com/gsoc2016/kvld.html
 ===Dictionaries===
 * <code>apertium-bel.bel.dix</code>: 105215 lemmata, 1807128 surface forms

@@ Line 68: / Line 68: @@
 !  Direction  !! Errors
 |-
-| '''Belarusian→Russian''' || 6.394
+| '''Belarusian→Russian''' || 6,394
 |-
 | '''Russian→Belarusian''' || 0
 |-
 |}
 ===Evaluation===

@@ Line 74: / Line 74: @@
 |}
 ===Evaluation===
+During the project next evaluations were completed:
+{|class=wikitable
+!  !! Size !! '''bel→rus''' WER (%) !! '''rus→bel''' WER (%)
+|-
+| 1 || 500 words || 59.19% || 55.63%
+|-
+| 2 || 500 words || 26.79% || 27.12%
+|-
+| 3 || 2000 words || 25.72% || 23.93%
+|-
+|}
+All texts and results (texts after MT and after postedit) can be found in [https://svn.code.sf.net/p/apertium/svn/incubator/apertium-bel-rus/texts/ apertium-bel-rus/texts].
 =Future work=

@@ Line 40: / Line 40: @@
 !      Coverage            !! '''Belarusian→Russian''' (%) !! '''Russian→Belarusian''' (%)
 |-
-| <code>'''Trimmed coverage''' (calculated in apertium-bel-rus)</code> || 84,3% || 83,6%
+| <code>'''Trimmed coverage''' (calculated in apertium-bel-rus)</code> || 84.3% || 83.6%
 |-
 !      Coverage            !! '''Belarusian''' (%) !! '''Russian''' (%)
 |-
-| <code>'''Raw coverage''' (calculated in apertium-bel, apertium-rus)</code> || 90,7% || 91,1%
+| <code>'''Raw coverage''' (calculated in apertium-bel, apertium-rus)</code> || 90.7% || 91.1%
 |-
 |}
@@ Line 68: / Line 68: @@
 !  Direction  !! Errors
 |-
-| '''Belarusian→Russian''' || 6,394
+| '''Belarusian→Russian''' || 6.394
 |-
 | '''Russian→Belarusian''' || 0

@@ Line 20: / Line 20: @@
 There were no other bigger changes in the Russian morphology.
 <br><br>
-The final version of the Russian morphology contains 84485 lemmas.
+The final version of the Russian morphology contains 84472 lemmas.
 ==Bilingual dictionary==
 The bilingual dictionary has been made from scratch. Different approaches were taken to generate bilingual entries, but main was translation with Belarusian-Russian dictionary<ref>[http://slounik.org/krapivabr Беларуска-расейскі слоўнік (Крапіва)]</ref>.<br><br>The final version contains 50,488 entries, but some of them were commented out to get clean testvoc in rus→bel.

@@ Line 4: / Line 4: @@
 As mentioned before, Belarusian morphological analyser was created from scratch. The main resource for importing words was dump of bnkorpus<ref>[http://bnkorpus.info/grammarDB.html Belarusian N-korpus – Grammar DB]</ref>. Unfortunately, there was available only outdated dump, that doesn't contain some frequently used words. These words were added semi-manually.
 <br><br>
-The final version of the Belarusian morphology contains 105213 lemmas.
+The final version of the Belarusian morphology contains 105215 lemmas.
 ===Adjectives===
 First problem is comparative and superlative forms for comparable adjectives. They are missed in bnkorpus dump and we didn't find any source for import. It was left to the future and now we have transfer rule for this case.

@@ Line 11: / Line 11: @@
 Also due to bugs in russian monolingual dictionary we were forced to exclude some verbs from analysis to get clean testvoc.
 ===Post-generation===
-For support "short U rule"<ref>[http://www.pravapis.org/fundamental_belarusian.pdf 1.8 Rules for в, у, and ў]</ref> were added additional 2-staged post-generation. Example:<br>
+For support "short U rule"<ref>[http://www.pravapis.org/fundamental_belarusian.pdf 1.8 Rules for в, у, and ў]</ref> was added additional 2-staged post-generation. Example:<br>
 <pre>$ echo "~уа ~уаа ~ба ~уа ~у ~уаа ~а" | lt-proc -p bel.pgen-short-u-1.bin | lt-proc -p bel.pgen-short-u-2.bin
 уа ўаа ба ўа ў ўаа а</pre>