File size: 31,791 Bytes
6fa4bc9 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 |
{
"paper_id": "W89-0123",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T03:45:45.382716Z"
},
"title": "Machine Aided Translation between the Two Norwegian Languages Norwegian-BokmM and N or wegian-N y norsk",
"authors": [
{
"first": "A",
"middle": [],
"last": "Bstract",
"suffix": "",
"affiliation": {},
"email": ""
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "The article describes essential parts of a prototype system for machine aided translation from Norwegian-Bokmal to Norwegian-Nynorsk. The central parts o f the system are a bilingual word list, inflection paradigms, phrases, and routines to deal with compound words. There are also syn-t2ictic and semantic rules, but they can be considered as preliminary. The article also includes a simple comparison of the two Izinguages in question. The program is written in TurboPascal and runs on IBM PC compatible machines.",
"pdf_parse": {
"paper_id": "W89-0123",
"_pdf_hash": "",
"abstract": [
{
"text": "The article describes essential parts of a prototype system for machine aided translation from Norwegian-Bokmal to Norwegian-Nynorsk. The central parts o f the system are a bilingual word list, inflection paradigms, phrases, and routines to deal with compound words. There are also syn-t2ictic and semantic rules, but they can be considered as preliminary. The article also includes a simple comparison of the two Izinguages in question. The program is written in TurboPascal and runs on IBM PC compatible machines.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "T h e m a in g o a l o f th e p r o je c t M a c h in e aided tra n sla tion fr o m N orw eg ia n -B ok m a l to N o r w e g ia n -N y n o r s k is t o a u to m a tiz e tra n sla tio n b etw e en th e tw o officia l N o r w e g ia n la n g u a g es, b o t h fo r p r o p o s in g tra n sla te d te x t, a n d fo r lo o k in g u p w ord s, p h ra ses a n d te x t p a rts. T h e r e is a lso a n im p o r ta n t o b je c t iv e to g a in k n ow led g e a b o u t m a ch in e tra n s la tio n in g en era l. T h e sy ste m w ill b e b a sed o n co m p re h e n siv e b ilin g u a l w o r d lists a n d lin g u istic rules im p le m en ted as d a ta fo r a c o m p u te r p r o g r a m .",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "T h e d e v e lo p m e n t h as s o fa r p a r tly b e e n s u p p o r te d b y th e N orw e g ia n R esea rch C o u n c il fo r S cie n ce a n d th e H u m a n ities (N A V F ) fo r 8 m o n th s in 1 9 8 7 -8 8 . T h e a u th o r o f th is p re s e n ta tio n has b e e n re s p o n s ib le fo r th e lin g u istic a n d c o m p u ta t io n a l w o rk . F o r m a l c o o p e r a t o r s a re a lso T h e N o re g ia n C o m p u t in g C en tre for th e H u m a n itie s a n d D e p a r tm e n t o f P h o n e tic s a n d L in g u istics a t th e U n iversity o f B e rg e n .",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Proceedings of NODALIDA 1989, pages 250-259",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "250",
"sec_num": null
},
{
"text": "T h e tra n sla tio n p r o g r a m a ",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "T h e Soft-\u00a3ind Hardw are",
"sec_num": "1.1"
},
{
"text": "T h e tw o oflScial w ritte n N orw eg ia n la n g u a g es a re th e m a jo r ity la n g u a g e N o r w egia n -B ok m 4 1 (N B ), w ith an unoflScial E n g lish n a m e D a n o -N o r w e g ia n , a n d th e m in o rity la n g u a g e N o rw e g ia n -N y n o rsk (N N ), w ith an u n officia l E n g lish n a m e N ew -N orw eg ia n . T h e s e la n g u a g es ca n , fo r foreig n ers, b e reg a rd e d as re la tiv e ly T h is la n g u a g e s itu a tio n im p lies th a t th ere w ill e x ist tw o N o rw e g ia n la n g u a g e s in th e fu tu re. A n d th is w ill m o tiv a te tra n sla tion to o ls b e tw e e n th e tw o la n g u a g e s, e sp ecia lly fr o m th e m a jo r ity la n g u a g e to w a rd s th e m ih o r ity la n g u a g e . the stoker/fireman has/isJn need for more coal) th e e x p re ssio n s in th e ta b le axe sim ila r fo r th e tw o la n g u a g es, e .g . th e w o rd s fo r g irls a n d h o u se s. T h is m e a n s th a t fem in in e p lu ra l in fle ctio n is th e sa m e (w ith s o m e e x c e p t io n s ), a n d th a t n eu ter p lu ra l in s o m e ca ses is sim ilar.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The Two Official Norwegian Languages, General Remarks",
"sec_num": "2"
},
{
"text": "In T a b le 2 th e re a re e x a m p le s o f g en itiv e , p a ssiv e v o ic e o f v erb s, co n g r u e n c e a n d n o m in a liz a tio n . ",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The Two Officicd Norwegian Languages, Simplified Comparison",
"sec_num": "3"
},
{
"text": "A t p resen t th e s y s te m is a b ilin g u a l tra n sla tio n sy ste m w ith tra n sfer rou tin e s, i.e. ta rg e t la n g u a g e d a t a is in serted in th e resu lt stru ctu res fr o m th e a n a ly sis o f th e s o u r c e la n g u a g e . A m o r e s o p h is tic a te d s o lu tio n , n o t im p le m e n te d y e t, c o u ld b e a n in te rlin g u a s y s te m , w h e re th e s o u rce te x t w o u ld b e tra n sla ted t o a la n g u a g e in d e p e n d e n t rep re se n ta tio n b e fo r e g e n e ra tio n o f th e ta rg e t te x t.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The Structure of the System",
"sec_num": "4"
},
{
"text": "T h e s y s te m is b a s e d o n s y n ta x a n a ly sis, a n d a s e m a n tic a n a ly sis g u id in g th e s y n ta x a n a ly sis. T h e d a t a files c o n s is t o f lin g u is tic ru les fo r c o n tr o llin g th e tra n sla tio n p rocess (i.e . a n a ly sis a n d g e n e r a tio n ) a n d refe re n ce d a ta fo r th e rules. F or th e tim e b e in g th e re is o n e ru le file, w ith s y n ta x ru les, a n d seven referen ce files, w h ich on a specific linguistic model, and the rules must be regarded as a basis for estab lishing a skeleton o f a syntax parser. Typical rules are ADNOM + NOM -> NP, V + NP -> VP, NP + VP -> S, PREP + NOM -> PP. The rules are at the time being based on the sequence o f SUBJECT -f VERBAL + OBJECT in main clauses, and ex istence o f inversion in certain types o f subordinate clauses. The syntax rules can to some extent handle relative clauses, conditional clauses and adverbial clauses.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The Structure of the System",
"sec_num": "4"
},
{
"text": "The context constraints concern both syntax and semantics. The syntac tic constraints concern the existence o f grammatical categories in the context, e.g. SUBJECT, OBJECT, 0BJECT2, THAT-clauses and INFINITIVES. The semantic constraints concern what semantic features may characterize the grammatical categories (see above) which is tied up to a kernel word or phrase (usually a verb). It is quite a big project to work out a com plete classification o f semantic specification for all words. Therefore our principle is to start with classifying words which otherwise may cause wrong analyses and translations. The parser uses these specifications to reject analyses where the semantic features are in confiict. If one or none o f the two phrases (words) have semantic specifications, a semantic check will not be performed. For further details, please see the de scription o f the data file containing the semantic classification.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The Structure of the System",
"sec_num": "4"
},
{
"text": "In th e b ilin g u a l w o r d list, a stem a ",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The Bilingual Word List",
"sec_num": "6"
},
{
"text": "T o g iv e s o m e im p ression o f w h a t th e sy ste m has t o w o r k w ith , I w ill p resen t a sim p lified c o m p a r is o n o f th e differen ces b e tw e e n th e tw o la n g u a g es, a n d c o n s id e r s o m e d ifferen ces in sp e llin g , in fle ctio n , sufiSxes, v o c a b u la r y , a n d sy n ta x . F irst it m u st b e s ta te d th a t th e tw o N o rw e g ia n la n g u a g e s h a v e m o r e sim ila r th an differen t e x p ression s, a n d th a t th e d ifferen ce s a re v e ry o ft e n sm a ll a n d co v e r lots o f ex p re ssion s in a s y s te m a tic m a n n er (e .g . d ip h to n g v ersu s m o n o p h to n g ).T h is reflects a c o m m o n h is to r y fo r th e la n g u a g es.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "Proceedings of NODALIDA 1989",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "T h e s y n ta x ru les axe b in a r y u n ifica tio n ru les w ith c o n te x t co n stra in ts .T h e sys te m is w o r k in g b o t t o m -u p , i.e. g r o u p in g re la te d w o rd s in la rg er eind la rger units u n til th e resu lt is c o m p le te sen ten ces. A t th e tim e b e in g th e rules are n o t b a sed",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "T h e sy ste m co n ta in s tw o d a ta files c o n s is tin g o f in fle ctio n in fo r m a tio n , a ta b le o f regu lar in fle ction p a tte rn s a n d a list o f irreg u la r in fle cte d w o rd s. T h e p resen ta tio n o f th e regu la r in fle ctio n p a ra d ig m s h as th e fo llo w in g fo r m a t fo r ea ch p a tte rn : p a tte rn c o d e -f-in fle ctio n e x p ression s. F o r o n e o f th e in fle ctio n",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
}
],
"back_matter": [],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "Selective Purism' and Language Auton omy. An \"intrarnational\" case",
"authors": [
{
"first": "Johan",
"middle": [],
"last": "Myking",
"suffix": ""
}
],
"year": 1989,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Myking, Johan. 1989: Term Harmonization, 'Selective Purism' and Language Auton omy. An \"intrarnational\" case. To be printed in the proceedings from The 7th",
"links": null
}
},
"ref_entries": {
"FIGREF0": {
"num": null,
"type_str": "figure",
"uris": null,
"text": "n d o th e r d a ta d e v e lo p m e n t p ro g ra m s (e s p e c ia lly fo r w ord lists) are w ritte n in T u r b o P a s c a l w ith o u t u sa g e o f g r a p h ic s emd s p e c ia l d a ta b a se p ro c e d u r e s , a n d is c o m p a tib le w ith v e rsion s 3 .0 , 4 .0 a n d 5 .0 . T h e s o u r c e c o d e takes a b o u t 50 k b , a n d th e c o m p ile d v e rsion a b o u t 34 k b w ith o u t th e d a ta tables. T h e p r o g r a m is ru n o n IB M c o m p a tib le m a ch in e s w ith th e o p e r a tin g sy stem D O S 4 .0 a n d low er. A m o n g th e m a ch in es is a T o s h ib a TIOOO w ith 512 k b R A M a n d o n e flo p p y d isc w ith 720 k b ."
},
"FIGREF1": {
"num": null,
"type_str": "figure",
"uris": null,
"text": "sim ilar b o th in sp e llin g , in fle ctio n , v o c a b u la r y a n d sy n ta x , w h ich w ill b e e x em p lified b e lo w . T o N orw eg ia n s th ese tw o la n g u a g es a re felt as d ifferen t, b o t h b e ca u se o f th e lin g u istic d iffe ren ces a n d b e c a u s e o f th e p o litic a l im p a c t o f th e lan gu ages. T h e tw o la n g u a g es h a v e b y p o litic a l d e cis io n b e e n s ta te d as o fficia l N o rw eg ia n la n g u a g es w ith e q u a l sta tu s. N N is b a se d o n N o rw e g ia n d ia le cts , w h ile N B is b a s e d o n a n o ld e r v e rsio n o f w ritte n D a n ish w h ich has la ter b een im p r o v e d w ith e lem en ts fr o m N o rw e g ia n d i a le cts. T h e r e is n o w a d a y s n o c o m p le te a g reem en t as t o w h ich o f th e tw o w ritte n la n g u a g es a re n earest t o N o rw eg ia n d ia le cts . T h a t d e p e n d s o n w h ich lin g u istic c a te g o rie s are co n sid e re d a n d th eir re la tiv e im p o r ta n c e . T h e r e is a ls o n o c o m p lete a g ree m en t t o w h a t ex te n t a w ritte n la n g u a g e sh all b e b a se d o n d ia le cts in o p p o s itio n t o w ritten tra d itio n ."
},
"FIGREF2": {
"num": null,
"type_str": "figure",
"uris": null,
"text": "o m p a r is o n is p re se n ted in tw o ta b les. T a b le 1 co n ta in s e x a m p le s o f sp e llin g , in fle c tio n , suflSxes a n d v o c a b u la r y . T a b le 2 co n ta in s e x a m p le s o f sy n t a c t ic c o n s tr u c tio n s . T h e e x a m p le s in T a b le 1 a re g r o u p e d in to fo u r co lu m s , tw o fo r ea ch lan gu age. T h e le ftm o s t c o lu m n b e lo w N B a n d th e rig h tm o s t c o lu m n b e lo w N N co n ta in th e fo r m s th a t a re m o s t d ifferen t fr o m th e o th e r la n g u a g e , a n d are u su a lly ca lled m o d era te fo r m s . T h e r ig h tm o s t c o lu m n b e lo w N B a n d th e le ftm o s t co lu m n b e lo w N N rep re se n t e x p re ssio n s w h ich a re re g a rd e d as a p p ro a ch in g fo rm s , u su ally ca lle d rad ical fo r m s . T h e sq u a re b ra ck e ts m a rk s fo r m s w h ich are a llow ed in w rit in g e x c e p t in d o c u m e n ts fr o m th e ce n tra l g o v e rn m e n t a n d te x t b o o k s fo r the p r im a r y a n d s e c o n d a r y s c h o o l, h ig h s c h o o l, a n d s o m e o th e r in stitu tion s. T w o o f"
},
"FIGREF3": {
"num": null,
"type_str": "figure",
"uris": null,
"text": "E a ch o f th e e x a m p le s c o n ta in s : N B p h rase, N N p h rase (p r e fix e d b y an a rro w : = > ) a n d an E n glish tra n sla tio n o fte n fo llo w e d b y som e g r a m m a tic a l in fo r m a tio n . W h e n th e w o r d o r d e r in o n e o f th e N orw eg ia n ph rases d iffers fr o m th e o r d in a r y E n g lish tra n sla tio n , I h a v e a d d e d a w o rd b y w ord tra n s la tio n o f th e p h ra se in th a t la n g u a g e . See fo r in s ta n ce M y k in g 1989 fo r an in s tr u c tiv e c o m p a r is o n o f th e tw o la n g u a g es. A sy ste m fo r m a ch in e a id e d tra n sla tio n w ill b e b a s e d o n th ese fa cts. T h e m o s t sig n ifica n t d iffe ren ce s a re fo u n d in sp e llin g a n d in fle ctio n o f w ord s. A s w ell, w e h a v e t o p e r fo r m a s y n ta c tic a n d se m a n tic a n a lysis (p a r s in g ) to id e n tify th e cu rre n t fo r m a m o n g h o m o n y m s (in s o u r c e la n g u a g e w h ich m a y im p ly differen t e x p re s s io n s in ta r g e t la n g u a g e ), t o en su re c o n g r u e n c e (w h ich ex ist in N N ; see e x a m p le s a b o v e ) a n d t o en su re re w ritin g o f s y n ta c tic c o n s tr u c tio n s ."
},
"FIGREF4": {
"num": null,
"type_str": "figure",
"uris": null,
"text": "T h e d e v e lo p m e n t is p e r fo r m e d b o t t o m -u p , i.e. it has b e en o f im p o r ta n c e t o e sta b lish a sk eleton o f a to t a l s y s te m , a n d th en reflne th e p arts. B e c a u s e o f th is, th e lin g u istic m o d e ls th a t axe u sed h ave t o b e co n sid e re d as p relim in a ry . T h e s y s te m c o n s is ts o f a c o m p u te r p r o g r a m a n d a c o lle c tio n o f d a ta files. T h e ru les a n d th e re fe re n ce d a ta is s to r e d in te x t files th a t ca n e a sily b e e d ite d . D u rin g th e d e v e lo p m e n t p h a se p a rts o f th e lin g u istic rules are p a rt o f th e p ro g ra m c o d e b e c a u s e o f p re lim in a ry d e sig n issues."
},
"FIGREF5": {
"num": null,
"type_str": "figure",
"uris": null,
"text": "are th e b ilin g u a l w o r d list, reg u la r in fle ctio n p a tte rn s , irregu la r in flecte d w ord s, p h ra ses, w o r d fo r m a tio n p a tte r n s (fo r c o m p o u n d w o r d s ), se m a n tic cla ssifica tio n , a n d in tern a l c o d e co n v e rs io n (w h ich w ill n o t b e e x p la in e d fu rth e r)."
},
"FIGREF6": {
"num": null,
"type_str": "figure",
"uris": null,
"text": "n d an in fle ctio n c o d e fo r b o t h N B a n d N N are req u ired . T h e fo ilw in g o p tio n a l ca te g o rie s axe presen t: c o n te x t c o n s tra in ts (fo r th e m o m e n t) fo r v e rb s (se e s y n ta x ru les b e lo w ) a n d s e m a n tic c la ss ifica tio n . T h e en tries a re stored in a lp h a b e tic a l o r d e r in a file w ith fix e d r e c o r d le n g th t o fa cilita te q u ick search es. T h e d a ta is u p d a t e d in a te x t file a n d c o n v e r te d t o fix e d file fo rm a t. T h e rep resen ta tio n o f ste m s are ca rrie d o u t a c c o r d in g t o a fo r m a t th a t en a b le s th e p r o g r a m t o sea rch fo r th e p a rt o f a w o r d th a t is c o m m o n t o all fo rm s , e.g . m u t are c o m m o n fo r m u tte r (in d e f. s in g .) a n d m u tre (in d e f. p lu r .), w h ich are th e N B w o r d fo r th e E n g lish w o r d nut. In sta n ce s o f d o u b le co n s o n a n ts w h ich are sin gle in in flected fo rm s a re rep resen ted w ith a h y p h e n in fr o n t o f th e s e c o n d co n so n a n t, as fo r in s ta n ce -t. A n d su ffixes w h ich a re re d u c e d in ce rta in in flecte d form s are p refix ed w ith a n a sterisk ( * ) , as fo r in sta n ce *er. A c c o r d in g t o th is th e en try fo r ' m u tte r ' is m u t -t *er. p a tte r n s fo r v e rb s , e.g . o n e o f th e tw o p a tte rn s fo r N B fo r th e v e rb k a ste (th ro w , c a s t ), th is m ea n s: v6 e er a a T h e r e a d in g o f th is is th a t ^lccording t o p a tte rn v 6 , th ese en d in g s sh ou ld b e a p p e n d e d t o th e ste m ka st t o p r o d u c e th e fo rm s k a ste (in fin itiv e ), k a ster (p re se n t te n s e ), k a sta (p a s t te n se ) a n d kasta (p a st p e r fe c t). T h e r e is a m o d u le in th e p r o g r a m w h ich uses th is p a tte rn in fo r m a tio n in c o m b in a tio n w ith the ste m fo r m a n d its in fle ctio n c o d e fo u n d in th e b ilin g u a l w o rd list. T h is rou tin e w ill b e e x e m p lifie d in m o r e d e ta il b e lo w . T h e p re s e n ta tio n o f irreg u la r in fle ctio n is d iv id e d b e tw ee n th e bilin gu a l w ord list a n d a list co n ta in in g th e irreg u la r fo r m s in N N . In th e b ilin g u a l w o rd list th e m o s t relevan t fo r m s (fo u r o f n o u n s a n d v e r b s ) o f N B are listed a c c o r d in g to th e a lp h a b e tic a l se q u e n ce . E a ch o f th e N B fo rm s are m a rk ed fo r a fo rm ca te g o ry in th e b ilin g u a l w o r d list, e .g . p resen t o f v e rb o r d efin ite sin g u la r o f a n ou n . In th e list o f irre g u la r N N fo r m s w e u su a lly fin d fo u r fo r m s o rg a n iz e d a cco rd in g t o a se q u e n ce , i.e. in fin itiv e-p resen t-p re te ritu m -p a st p e rfe ct, th a t expresses w h a t fo r m it is."
},
"FIGREF7": {
"num": null,
"type_str": "figure",
"uris": null,
"text": "s e s a re s to r e d in a te x t file t o w h ich th e p r o g r a m has a ccess. W o r d s th at u su a lly o c c u r in th e p h ra se in o n ly o n e fo r m are p resen ted in th a t fo rm in th e file. W o r d s th a t m a y b e in fle cte d as p a rt o f th e p h ra se a re p re se n te d in in flected fo r m s (a t th e tim e b e in g in differen t e n tries, b u t th is w ill s o o n b e c o n ce n tra te d in t o o n e e n tr y ). A w ild c a r d (\u25a0\u2022\u25a0) h as b e e n in se rte d in w h ich an a d d itio n a l w ord c a n b e in se rte d (th is w ill b e e x p a n d e d t o p h ra se s), e .g . a d v erb ia ls as fo r in sta n ce n o t. E a ch p h ra se (r e c o r d ) is p refix ed w ith o n e o f th e w o rd s in th e ph rase. T h a t w o r d a c ts a s a k ey t o th e p h ra se. T h is m ea n s th a t th e k e y w o rd is a lso listed in th e b ilin g u a l w o r d list w ith a c o d e th a t tells th e p a rser t o lo o k in to th e phrase list. T h is ke}Tword is u su a lly ch o se n a m o n g w o rd s th a t is reg a rd ed as h a v in g low fr e q u e n c y in t e x t, t o p rev en t t o o m a n y su p erflo u s search es. T o th e righ t o f th e p h ra se w e fin d th e N N s y n o n y m , p re fix e d w ith a d o t ( ' . ' ). A n e n try m a y lo o k lik e th is (vO is th e in fie c tio n c o d e fo r irreg u la r v e r b s ): behov ha vO * behov for .trenge vO 9 W ord Formation, Compound Words A m o d u le h as b e e n im p le m e n te d fo r th e a n a lysis a n d tra n sla tion o f c o m p o u n d w o r d s . T h e m a in p rin c ip le is th a t c o m p o u n d w o rd s th a t d o n o t e x ist in th e b ilin g u a l w o r d list m a y b e a n a ly sed as b e in g c o m p o s e d o f w o rd s in th e list. T h e p r o c e s s in g o f w o r d fo r m a tio n h as tw o im p o r ta n t ty p e s o f restrictio n s. O n e t y p e o f re s tr ic tio n is th a t w o r d s th a t a re p a rt o f a c o m p o u n d w ord m u st u su a lly b e th e s o -c a lle d fo r m 1, i.e. in fin itiv e o f v e rb s, in d efin ite sin gu lar o f 256 Proceedings of NODALIDA 1989 n ou n s an d c o m m o n g e n d e r (i.e . m a scu lin e a n d fe m in in e ) o f a d je c tiv e s . B u t th e rig h tm ost (la s t) p a rt o f th e w o rd m a y o f c o u r s e b e in fle cte d . A t th e tim e b e in g th ese re strictio n s are in clu d e d in th e p r o g r a m c o d e . T h e o th e r ty p e o f re strictio n c o n c e r n s w h a t p a rt o f s p e e ch th e d ifieren t w o r d p a rts ca n c o m e & om . A n a d je c tiv e a n d a n o u n m a y b e c o m b in e d , lik e r\u00f8d v in (re d w in e ). B u t w e c a n n o t c o m b in e a n o u n a n d a n a rticle , like g u tte n (t h e b o y ) read as g u tt + e n (b o y + a ) a n d w ill b e tra n sla te d t o g u ttein . In th e s y ste m th ese ty p e s o f re strictio n s are w ritte n in to a d a ta file w h ere p a tte rn s ex p re ss illegal se q u e n ces o f p a rts o f sp eech ."
},
"FIGREF8": {
"num": null,
"type_str": "figure",
"uris": null,
"text": "A s m e n tio n e d a b o v e , a m o d u le fo r sem a n tic cla ss ifica tio n t o g u id e th e p a rse r has b e e n in clu d e d . T h e sy ste m is im p le m e n te d as a th esa u ru s in a tree s tru ctu re . T h e raw d a ta is w ritte n in to a te x t file w ith e x p re ssion fo r th e s u b o r d in a te c o n c e p t in th e left co lu m n a n d th e e x p re ssio n fo r th e s u p e rio r c o n c e p t in th e rig h t. FVom th ese d a ta th e p r o g r a m b u ild s th e th esa u ru s stru ctu re . T h e p a rser u tilizes th is in fo r m a tio n t o d e c id e i f tw o w o rd s o r p h ra ses ca n b e unified a c c o r d in g t o th e sem a n tic c la ss ifica tio n . T h is is o f im p o r ta n c e w h en th e tw o w o r d s /p h r a s e s h ave s e m a n tic re p re se n ta tio n a t d ifie re n t levels in su ch a w a y th a t o n e o f th e rep resen ta tion s is s u b o r d in a te d t o th e o th e r. T h a t m ea n s th a t th e y are co m p a tib le . 11 The Morphological Analysis for Single Stem Words T h e m ain c o m p o n e n t o f th e w o rd a n a ly sis is r e c o g n itio n o f w o r d s in th e b ilin g u a l w o rd list a n d th e rela ted in fle ctio n en d in g s in th e in fle ctio n ta b les. In th e a n a ly sis o f c o m p o u n d w o rd s th is in fo rm a tio n is c o m b in e d in r e p e a te d w o r d -r e c o g n itio n d u rin g th e w h o le c o m p o u n d w ord . T h e g en era l s tr a te g y fo r w o rd r e c o g n itio n w ork s fr o m rig h t t o left. T h e p r o g ra m first sea rch es in th e w o rd list fo r th e w h o le w o rd . I f it is n o t fo u n d , a n ew search is p e r fo r m e d w ith th e rig h tm o st ch a r a c te r d e le te d . T h e sea rch p r o c e s s is rep ea ted w ith d e le tio n o f th e rig h tm o st ch a r a c te r u n til th e re w ill b e a m a tch o r th ere w ill b e n o sea rch strin g left. I f th e in p u t w o r d is k a s te r th e p r o g r a m is a v e r b stem in th e list. K a s t has tw o d ifieren t in fle ctio n c o d e s , v 6 a n d v7. T h e p r o g r a m w ill e x p a n d th e stem w ith en d in g s fr o m b o th th e in flection p a r a d ig m s in o r d e r t o fin d id e n tity b etw ee n th e c o m p le te in p u t w o r d a n d o n e o r m o r e in fle cte d fo r m s o f th e w o r d list en try. T h e p a ra d ig m s are: o r d in g t o th is th e ste m k a st a n d in fle ctio n en d in g s p r o d u c e th ese w o rd form s: s e c o n d w o r d fo r m s , i.e. th e p resen t te n se in b o t h th e p a ra d ig m s m a tch . T h a t m ea n s th a t th e p r o g r a m h as id en tified k a s te r as p resen t ten se o f th e stem kast."
},
"FIGREF9": {
"num": null,
"type_str": "figure",
"uris": null,
"text": "h e a n a ly sis o f c o m p o u n d w o r d s fo llo w s th e sa m e m a in stra te g y as fo r sin gle s te m w o r d s . T h e m a in d ifieren ce is r e p e a te d sea rch es fo r stem s, e n d in g s a n d jo in in g ch a ra cte rs. I f th e in p u t w o r d is lae r e r h \u00f8 y s k o le s tu d e n te r (T e a c h e r s ' T ra in in g C o lle g e S tu d e n ts, rea d stem b y ste m as: ' te a ch e r + h ig h -I-s h o o l + stu d en t + s ' ) th e p r o g r a m w ill d e le te th e rig h tm o st le tte rs u n til it m a tch e s l<er. T h a t m eans th a t sea rch es w ill b e d o n e fo r th e fo llo w in g ch a ra cte r seq u en ces: lae re rh \u00f8 y s k o le stu d e n te r lae re rh \u00f8 y s k o le stu d e n te lae re rh \u00f8 y s k o le stu d e m a tc h e d stem s in th e w o r d list a re th e n o u n stem lae r w ith th e suflSx -e r a n d th e tw o m a scu lin e in fle c tio n c o d e s m 3 a n d m 4 , a n d th e v e r b stem lae r w ith th e in fle c tio n c o d e v l . T h e p r o g r a m first e x p a n d s th e se tw o stem s w ith th e th ree p a r a d ig m s , like th is: e o f th ese w o r d fo r m s m a tch th e w h o le (c o m p o u n d ) w o r d lae r e r h \u00f8 y s k o lestu d en ter. T h e n w e h a v e t o a ctiv a te th e m o d u le fo r a n a ly sis o f c o m p o u n d w o rd s. A s sta te d a b o v e , a c c o r d in g t o o u r ru les, w o r d s w h ich a re p a rt o f c o m p o u n d w o rd s ca n o n ly o c c u r in fo r m 1. In th is ca se th ere a re th ree c a n d id a te s th a t m a tch es th e b e g in n in g o f th e c o m p o u n d w o r d , o f w h ich tw o w o r d fo r m s a re u n iq u e: Itsrer an d lae re. T h is m ean s th a t th e rest o f th e w o r d is e ith e r h \u00f8 y s k o le s tu d e n te r o r rh \u00f8 y sk o lestu d en ter. T h e ])rogra m tries b o t h strin g s. It fails fo r th e la st o n e , b u t w ill su cce e d fo r th e first. T h e id en tified w o r d fo r m s as p a rts o f th e c o m p o u n d w o rd are h \u00f8y, sk o le a n d stu d en ter. T h e p r o g r a m w ill a ls o p r o p o s e th e n o u n s tu d e n t an d th e c o p u la v e r b e r in stea d o f stu d en ter. T h is s o lu tio n w ill b e e x c lu d e d b e c a u s e o f th e rules w h ich im p ly th a t p a rt o f sp e e ch seq u e n ces a re n o t a llo w e d in c o m p o u n d w ord s. T h is w o r d w ill b e tra n sla te d t o N N lae ra rh \u00f8g sk u lestu d en ta r. 13 PcLTsing Strategy T h e p a rsin g s tra te g y is b o t t o m -u p . It is b a se d o n b in a r y rules like th o s e liste d un d er \" S y n ta x ru les\" . T h e ru les co n sists o f tw o c o n jo in in g (a p a ir) s y n ta c tic la b e ls a n d re striction s fo r th eir s y n ta c tic fea tu res (g e n d e r, d e fin itn e ss, n u m b e r, ten se e t c ). T h e p a rsin g m o d u le p e r J u n e 1989 is a p re lim in a ry o n e . T h e d eta ils o f th e sy n ta ctica l a n d sem a n tica l d e s c r ip tio n a n d p r o c e d u r e s w ill b e r^ldicaily rev ised an d a c c o m o d a te d t o a m e t h o d o lo g y b a se d o n a L e x ic a l F u n ctio n a l G ra m m a r (L F G )."
},
"TABREF1": {
"num": null,
"content": "<table/>",
"text": "",
"type_str": "table",
"html": null
},
"TABREF3": {
"num": null,
"content": "<table/>",
"text": "",
"type_str": "table",
"html": null
}
}
}
} |