Commit Graph

2227 Commits

Author SHA1 Message Date
aarne
00a789d624 fixed "date" and "force" in DictFin 2013-03-29 21:46:18 +00:00
aarne
660d21000d Some more DictFin items. 50 of the 3220 Penn examples remain to be covered. 2013-03-29 21:34:50 +00:00
aarne
5c75ec2cd2 top-ten missing DictEngFin words, now translating 95% of the corpus 2013-03-29 20:51:30 +00:00
aarne
9414503235 a few more words in DictFin, but the most frequent missing ones - now complete lin for over 90% of the complete trees 2013-03-29 17:42:47 +00:00
aarne
01f12a135b preliminary implementation of GerundN and GerundAP in ParseFin. More coverage, but some strange translations. 2013-03-29 12:45:24 +00:00
virk.shafqat
d71d5ae25b added around 3K new entries in DictHin
Place the long patch description above the ***END OF DESCRIPTION*** marker.
The first line of this file will be the patch name.


This patch contains the following changes:

M ./lib/src/hindi/DictHin.gf -25709 +28154
2013-03-29 11:34:14 +00:00
aarne
ed570d37e6 removed unstemmed DictEngFin and ParseFin: they were no more maintained and would not scale up anyway 2013-03-29 10:21:29 +00:00
aarne
ed35291768 updated log.txt of ParseEngFin experiment 2013-03-29 10:20:36 +00:00
aarne
eb791244bd some new opers in ParadigmsFin, and 200 more words in DictEngFin: out of 3220 Penn trees now 2721 are completely translated (but mostly not so well...) 2013-03-29 10:13:04 +00:00
aarne
27a1811446 second round of completing ParseFin with missing constructs: hundreds more translations 2013-03-29 08:03:20 +00:00
aarne
2eab210435 new DictEngFin in finnished/stemmed. Only 9k lemmas missing. 2013-03-28 14:22:09 +00:00
aarne
abd95ff3c1 preparing a new DictEngFin from WN data 2013-03-28 11:04:09 +00:00
kr.angelov
ad4c97fdf7 added a few more multiword expressions in DictEng and a few words in the abstract syntax are not tagged with their senses. There is a new statistical model too 2013-03-27 20:46:42 +00:00
aarne
26175a69df mapped DictEng to Fin via Wordnet and Kotus 2013-03-28 08:16:45 +00:00
inari.listenmaa
80f4d1de2e Pronouns with possessive 'llur'
Changed default plural possessive pronoun from llur to "el seu / la seva" etc. and created variants with "llur / llurs" in ExtraCat.
Also changed good_A in LexiconCat to a preceding adjective (prefA).
2013-03-26 16:45:11 +00:00
aarne
a01ac1370d linking Kotus with FinWordnet 2013-03-26 15:11:59 +00:00
aarne
379647e5d9 a Haskell script for looking up words and compounds in Finnish Kotus 2013-03-25 16:50:51 +00:00
inari.listenmaa
af53e6e31b Corrected VRefl in DiffCat
Fixed VRefl in DiffCat from "se" to "es" and tonic from "sÌ" to "si". Also some trivial formatting things in MorphoCat and IdiomCat.
2013-03-26 11:33:54 +00:00
aarne
2bb9eb5d14 the path to PROBSFILE in lib Makefile 2013-03-25 14:28:12 +00:00
aarne
eb9685e911 added generalized ComplVV to ParseFin 2013-03-25 14:21:46 +00:00
aarne
77a9530cf9 fixed DictHin.quite_Adv which had wrong type 2013-03-19 08:09:31 +00:00
kr.angelov
72556ad1ae a long list of prepositions from Wikipedia is now imported in DictEng in addition there are a number of small other changes in the dictionary. The statistical model is updated and is now moved to treebanks/PennTreebank/ParseEngAbs.probs 2013-03-25 10:24:24 +00:00
kr.angelov
466813f1e8 fix in ParseHin which made it impossible to load the grammar with the C runtime 2013-03-20 10:34:37 +00:00
virk.shafqat
648b3c53a9 pathbugfixParseHin 2013-03-20 10:21:12 +00:00
virk.shafqat
af4ac56b35 smallfixDictHin 2013-03-20 09:17:55 +00:00
kr.angelov
8041999405 the ParseFin grammar now excludes ComplVV from VerbFin since this function has a more general type in the parsing grammar 2013-03-19 10:49:13 +00:00
aarne
ed8611c94b Makefile entries for ParseHin and ParseGer 2013-03-19 07:59:04 +00:00
aarne
6660a47404 Makefile entries for pgf's for robust parsing 2013-03-19 07:26:06 +00:00
kr.angelov
c775d0c5c5 filterout all adjectives and adverbs which could be derived morphologically 2013-03-18 17:31:20 +00:00
virk.shafqat
4fd9c860f3 factorUrdandAddDictGer 2013-03-18 14:42:10 +00:00
kr.angelov
34fddf669f some of the newly added nouns in DictEng were actually variations of already existing lexical entries. Those are removed now. 2013-03-15 23:23:06 +00:00
kr.angelov
e5913189db massive extensions in DictEng and DictEngBul. This includes all new nouns imported from WordNet by Shafqat, phrasal verbs that I collected from internet and the PennTreebank, plus various other small additions. 2013-03-15 20:18:22 +00:00
aarne
ff879641c8 uncomment a VerbFin function left commented by mistake 2013-03-15 17:40:28 +00:00
aarne
23a310be57 a stemmed version of DictEngFin and ParseFin - now works without trouble! 2013-03-15 15:07:14 +00:00
inari.listenmaa
2295f491c7 Corrections in LexiconSpa
Corrected many verbs in LexiconSpa. Also added in BeschSpa oir_51 and corrected morder_50b. In MorphoSpa added words ending in é and á to a regexp for regular nouns.
2013-03-15 17:13:25 +00:00
aarne
87d465594b stemming in Fin also for verbs now 2013-03-15 11:13:49 +00:00
aarne
3b66a870d1 stemmed version of verbs: from 102 forms to 13 "technical stems" in Fin 2013-03-15 08:55:14 +00:00
aarne
330a2ae094 Stemmed Finnish for N and A. Cuts down the number of forms stored in the lexicon to 30%, requiring more often -bind in linearization. Stored as alternative files in finnish/stemmed; perhaps to be functorized later. 2013-03-14 20:55:59 +00:00
aarne
59cbd2867f experimental stem-based morphology for Finnish, to minimize lexicon size, started 2013-03-14 15:13:44 +00:00
aarne
54324d5de3 added PassVPSlash to ExtraRomance; it is a generalization of PassV2 already implemented for some languages. As a variant in Spa and Cat, added PassVPSlash_ser with another copula. 2013-03-13 17:07:57 +00:00
aarne
18324e5e04 Poss,Part,Count NP implemented in Finnish: tehtaan johtaja, lasi viiniä, kolme pojista 2013-03-13 14:48:23 +00:00
aarne
b3e1908c78 Poss,Part,Count NP in Scandinavian 2013-03-13 14:21:26 +00:00
aarne
31b4b13b65 constructors PossNP, PartNP, CountNP to separete "of" constructions 2013-03-13 13:42:41 +00:00
aarne
05c9934af0 added the nominal possessive form NPNomPoss to NPCase in Eng (covering mine, yours, etc) 2013-03-13 12:34:22 +00:00
aarne
f61a013a40 case endings in Fin symbols: Nom is a special case, since there is no ending 2013-03-12 20:29:31 +00:00
aarne
367123c81b the proper place of verb particle in Eng passives 2013-03-12 20:06:51 +00:00
aarne
d14589804e worst case of Fin mkA; default case endings for Fin symbol as PN 2013-03-12 20:01:07 +00:00
aarne
77171607a9 some additions to DictEngGer, e.g. said_Quant 2013-03-12 16:59:38 +00:00
aarne
c5d03231d8 removed variants from IrregGer, introducing extra fun's for irregular variants, but none for regular ones. 2013-03-12 16:21:37 +00:00
aarne
9cd632497b AdvVPSlash in Fin 2013-03-12 10:00:15 +00:00