開発課題の考察、技術用日本語の検証実験、市場
Contents
1.2.
3.
4.
5. 3 1 1 3 1 1
6. NE
7. ee
8. Alc 4S BADE Sen oO AAA Hr
9. Web 2
10. E 3 2 1 0 0
11. re HARTI BAAR OTIC RT ODE BIA
12. Sul BLT Vee git a yt
13. 2 NN tt
14. in eR MLIB IE 1 1 3 1 2 3 1 2 1 eal aay MSR aa M0 SER SER REROLIBAE EE AB PLE 1 Nie nie etre NW
15. ul 3 CDL CDL CDL Concept Description Language 3 1 1 CDL
16. AGE OPER AT DEBATES OLD CES XM 3 2 1
17. CDL CDL CDL
18. 2 os NN AM lt
19. qd ot NM CDL lt gt lt gt lt gt lt gt lt gt lt gt lt gt lt gt lt gt lt gt lt gt gt
20. ae Oe a J
21. 3 1 2 So 6000 2007 1 Se 2000 4 1995 1991 12 11 18 17 1987 125A 38 27 2003
22. OE 2B 2 0 3 4 28 TEL 03 3454 1311 1 7 TEL 03 3615 5511
23. LY 3 4 2 E 1 168 20 230
24. 0 gt Of J 2008 3 0 Xt 1
25. CDL Web 2 3
26. 2001 oa OF Nae 2 1 2 3 3 gt ER 20 1 BORD b 1 H
27. Simplified Technical English STE EU23 IBM
28. E a ee a oa E v
29. 2 1 2
30. we 2 45 0 1 tm
31. 1 2
32. 12 2 3 13 SVO S O S V O O V S Ex 3 3 E 3 2 1 1 2 3 3 E 3 2 1 1 2 3 14
33. O ee SEY ChB
34. 3 3 3 3 4 3 4 1 Te C
35. 4 2 ee Oe eee XML RDF 4 3
36. A 3
37. x 3 3 6 E argmax lt E gt P E P J E o P E P E pil E E OH Ree
38. 3 3 7 3 3 8 3 3 7 gt pa 1 O X
39. 01 SSM 02 0 Oe ee ee OR IM lacs OO 05 ag lt i ob 06 eS os cae META i sagas 09 SS EY i ee 10 ope 1 ae ood 7 ICT av 12 a 7 lt Sealey 13 V Be Foe 14 Q amp A
40. EC 2 0 3 1990 2000 Web
41. 8 HL 1
42. F BORO MEE ML SES CLICK OC at cz We gt Aided Patent Engineering Sh E
43. Cr im II lt gt 3 5 2 1 AK QA AU HES 5 G Ee Bel SIAL FY CRE Se BA Se At AY Bat YASH L6 4x ods 1 0 LAH se OU TAY ASAE BEN A YE 3 9 SS
44. gt gt SE Aan EN ai i F f ean ee tee fe 2 7 CA PE Computer F ARO 6 Phas E
45. CDL HOORAY AF ADE LV lt 3 2 5
46. TLO ME ete ray HE
47. A AA 2
48. 2 0 3 3 ee BRIER K de RF A OE RE EE EE 1 2 8 SR a i 2 De BD ORE EER OE CE CO Re 5 600 ep I oe SP Ny a EE 5 9 Axes O Bere CO eR es eee tenes ban Dia i 5 13 Ss NR 15 3 2 0 17 SPD NN Aetas 22 3 2 3 ki 27 3 3 iii 32 3 3 DD 32 3 3 2
49. CDL CORI LISD Sul Of
50. 8 2 4 3 2 3 1 9 5 Hy 20 OC HRT SO AAT O REL FAME LC BOA
51. 8 0 1 2 ao 1 3 3 3 9 3 3 3
52. O x i S 8 cv lt 1 X ul ae E argmaxEP E P JIE is obtained by applying Bayes theorem to this formula P E of this formula is called a language mo
53. r ao
54. SN MEN FEA ZS A AEB D gt BART Ze 20 oS A
55. 1 1 4 3
56. 3 3 1 7 2 CDL 2 3 7 7
57. Ue lp es D Fi ak AS Sh A TP
58. 3 1 1 3 2 4
59. 4 15 Ex XML Web XML Web XML Web by Ex on Ex
60. gt 3 3 9 a a ea ll Ee ET EE CZ re Nh 8 St Fe SO RS SH Pe NO 8 St 2 tS bi St D H SF et aH _ ac iz 1 am ma a wed oo vo LL LLL ea PI Fresh onl oslo SC i iz i a 119 100 12g LO 72 87 79 28 74 36 99 53 x 9 17 5x tex 10 se of io ed ol enol ons ao mes 5 aya Fitbo TOME 3 3 10 3 7 Hl 72 OO 3 x
61. 3 2 2 2 Web he
62. _ B oe B FENDER IT LT IRD gt K B uValue B B RK CEE Siete de a Fe
63. CDL 2 CDL a
64. 2 NN 5 MN Me 3 1 2
65. 4 1 7 amp LAG semantic authoring Ke yt Hae HEY EY 1 1 1
66. i 7 2 1 Tes
67. b c d CN oe ne
68. Kaw iii ii SIU DAT gt ai LEO HE i E Web Hak ROR
69. CDL Concept Description Language CDL OR EC HS 10
70. a b f oF
71. TEs 3 5 1 WP MT 4 230 CDL 3 SS V0 CDL CDL ISeC Japio 3 5 1
72. O O 3 3 2 3 3 4 6 3 3 4 MUO MAO AMMO
73. 7 5 3 6 9 3 3 10 2 i ef owl of oof _ a of maa vax 205 00 poner gl ela gl pal in BSH yt x Said Sai aS et EA S BH At Dat amp BU aS et ho Ng Se 2 ae ae 7 i 26 17 Tt 49 ox x Ee seo ll se se se DI ae w ae Co ae Q ae ae Co ssf o a ET LI 98 3 0 58 10 0 0 0 3 12 3 17 onl ml on me 78 6 6 2 10 lr ololaomlr SRI N Re a ae se se PTET TET TP we Co ae Co i 2 se Le 80 80 J a ae xlolel lolel xlole Nlo ololelm ae ed ee o 16 ae ae seis 1 x 9 19 Q Es ss se 3 3 3
74. E CDL Sh F RL Fark EFA LBA BIT Zt SCE 2 FRI SD Te D SCHR GR BE E
75. 22 6 Ex 23 7 Ex CDL 4 4 4 4 1 2
76. Vv P J E 3 P J E P J E
77. 3 1 1 1 LU CORSIE
78. 3 3 9 7 5 3 9 6 41 53 Xx 6 7 4 7 Xx 6 7
79. E VAF LBV 4 ALIVE ee i ZB RAK ZA 3 eis CEL OT FA AKL CWS ROE E A Vek Ole E
80. oe 2 P J E P J E 2 E 1 2 E 1 J J O J PE P PE P J E SES pees et O J 5 F 2 E 3
81. H lt gt lt gt lt gt lt gt lt gt lt gt lt gt 5 1 lt gt lt gt lt gt gt lt gt lt gt Bil lt gt lt gt CE lt gt
82. 3 3 1 2 3 1 Ex Ex G ny 2 Ex
83. a 2 0 0 6 2 0 0 7 amp at FF Be Bull
84. K Se ar Da A ai ay um rH in cH 3 1 9 1 0 2 0 2 4 3 1 3 2 BRO SS AS 3 3 3 4 3 1
85. Aa te SHER 3 11 Be NBS BNE BID HD BHERET IL 15 BHERET IL 18 4 CTA RHER CISL PHAR TL AU CHIR ETE 9 1 2 4 1 2 3 3 2 AL
86. I 3
87. 2 nee EEE ERO FB 72 EADS AIT BU IX1 X2 Xn
88. a b oe CDL Concept Description Language CDL jpn surface 1 3 CDL
89. MPN CDL i CDL CDL DRAFT 4 T amp OAM ZEEE DIT AS OE BE FF aN ad
90. 1 RAS 1 324 2003 13 22 44 60 Hy ul 200 3 4 3
91. lt 16 3 17 3 18 dae Ex 59 29 6 5 1 5 9 2 6 5 1 Ex 19 E 20 Ex 21 5 Ex
92. ul vy re v Be aT ee AA FB REESCERIRER A source language and the purpose language of In translation model creation the source translation are called a target language for the language of translation is called a source language and the purpose language of translation ed a target language 6 source language of the translation in the case of translation model creation is cal x 3 3 1
93. gt CDL jpn eng chi CDL math prog movie music CDL core RDF OWL CD a CDL 4 b A E c CDL d
94. 36 3 3 Ramee PN Ga a chee BS iets ED ae MS SE ae ee OT ea 44 ee coe ye A Aca le 46 3 4 vigituey LAN se 46 a eee ML 9c ee 209 a ee TEE vr 46 3 3 PU TC A Se EAs asstssatandensnssasansdnonhandaradsnnsananeeddorsavauadaananave 47 ON RN 48 4 Ek Rk 51 A A eer ae uc EE 51 4 2 51 a Py pe ed acd Be aie ed nies MN I 52 1 PAA ICBVY CIA 1978 Plain English
95. SQL c CDL a c b b b CDL CDL CDL nl CDL for natural language CDL nl a ee ae 3 2 6 CDL CDL AUN VAS av CDL nl amp jpn CDL
96. alana be NPO SeC W3C Web Language CDL Concept Description 3 a Aan b
97. ls an 0D 3 3 EL 4 DF 1 5 S1 S2 1 6 7
98. E argmax lt E gt P E P J 1E E argmaxEP E P JIE is obtained by applying Bayes theorem to this formula P E E Inside P E of this formula is called a language model and shows the appearance likelihood of the word in the sentence of target language E 7 d E argmax lt E gt P E P J 1E E argmaxEP E P JIE is lobtained by applying Bayes theorem to this formula P E E P E E Inside P E of this formula is called a la
99. 11000 CD ROM1 1997 9 ae NTT 80000 30 1999 9 63000 14000 CD ROM CD ROM 2003 2005 3 Readability
100. 25 7 6 9 72 3 2 3 8 2 5 9 X 69 16 5 3
101. r TEA 16 SeC LL KA VAT LD EB AREA A ie ie B K A ESE
102. 3 2 2 1 3 OD a JEAN ZR A bh RAZR A ARAL ERK BEIE b amp KNIT ATA SISH MOE ICAAMER F ey 7 L L NDB VBA SETS PRUS HAMITRT Ah EMERTSLEXOA VIUYTA c
103. 1 i1 D 2 i 2 dO dO D2 D2 i1 i i2 SIE COD 0 Hi izo 8 Bs CI D2 1i2 i1 2 WL cans RTC SADR M1 Ea C
104. 3 3 1 3 338 3 3 2 oul 3 3 2 Tul Tt _J_ 47 53 69 80 12 7
105. LOS SL 2 2 0 3 KEIRIN aes ia http ringring keirin jp ar
106. 2 7 3 3 5 3 3 6 Tul Tu 3 3 5 x
107. 4 1 f ee 4 HE
108. B FL BB B URIS ik B is B CI ieee
109. 3 2 7 3 3 3 1 2 3 3 3 1 CDL 2 0 7 2 3 CDL a can NON age een rd NII at 7 2 3 ON Hml
110. F 61
111. n 2 8 9 10 11 Ex 1
112. 21 23 3 2 1 ie ong ada
113. CDL bt
114. Ex Ba Lys CDL E R 2 3 CDLjpn surface a FAMI SCHEER TSE OICSFAREBER GAFAVTZ b
115. THE DAILY _ _ 1990 11 1 YOMIURI 17 Wikipedia KY CHILDES NICT RE JCHAT 661 2005 5 6 1 25 752 120 6 2001 5 462313 2008 1 20 48 000 B5 4 1 1996 1998 3 ft 73 2003 12 Tt 31 90 23MB
116. 0 1 ot i 1 3 3 3 0034 2 1 1 er 1 RISB OF VIOU lt BA
117. at Sh an Y ea DAE 4 ERE OE OF AL 2 FE MASSA ae at aot ive PN 2 R F H E
118. NR 9 S OS TAX AMBRE A BES HEY SY T O wo Sa TLL ANE hia feayse Lae RAR My Bg Fd Fy Coy AER LAE AHO 9 Sia 20 serene ee Pass ee ls x fel SSP IY BEE By WF By Het seem rey cee Med TEX AKA ft EREY BE Td THK LY KELLEY AS g Seb y 6LH sv otdeP D39SI HIANYW A TG0 1d0 5 V MOTE 0 MOTE IA MIB ME IME YMA TV LOK ESE BEEN AWE Bla gH lay 8ZH ZZH Bla IZH 0ZH 61H ME Rl IRE ae BAL ts Ea fee HAW 3 9 8 4
119. a4 ef EA AR 0 0 Ls Kl 3 2 3 3
120. jpn surface CDL jpn surface CDLjpn surface CDLjjpn surface CDL jpn surface
121. 3 2005 000031 7 344 2005 003747 20 714 2005 004500 8 549 2005 004707 1 147 2005 010347 12 351 2005 025312 5 470 2005 025474 8 113 2005 025653 15 303 2005 025659 13 258 92 249 E argmax lt E gt P E P J E P E E a a b c
122. OV CIL SER CREE LE LB HAA EEO
123. del and shows the occurrence likelihood of the word in the sentence of target language E called a language model and shows the appearance likelihood of the word in the sentence of target language The latter P JIE is called a The latter P JIE is called a translation model and the translation model and probability that sentence J oflexpresses the probability the 1st language will be that sentence J of the 1st generated from sentence E language will be generated of the 2nd language is from sentence E of the 2nd expressed language Conditional probability P E P J E mentioned above to input sentence J generates translation E used as the maximum using this language model and a translation model Translation E whose conditional probability P E P JIE is the maximum to an JL input sentence is generated using this language model and a translation model 4 3
124. he sentence of target language E P E of this formula is called a language model and shows the occurrence likelihood of the word in the sentence of target language E E argmax lt E gt P E P J 1E P E E P E E O min 2 BR x ees
125. nguage model and shows the appearance likelihood of the word in the lsentence of target language E E argmaxEP E P JIE is SBIZKY E argmaxl lt E obtained by applying Bayes gt JP E P J E theorem to this formula sos sex Inside P E of this formula is P E of this formula is called eae called a language model and lalanguage model and shows E E shows the appearance the occurrence likelihood of DUBERESTLOCHE a likelihood of the word in the Ithe word in the sentence of mem eA 28 sentence of target language E target language E ALE AT LO SPAR E argmaxEP E P JIE is lobtained by applying Bayes theorem to this formula Inside P E of this formula is called a language model and shows the appearance likelihood of the word in t
126. ontrolled English English H CHO OV AF AIH aes ff i CPL Computer Processable Language Standard Language Simplified Technical i Controlled English EU 2 SEKT Semantically Enabled Knowledge Technologies amp U ACE Attem
127. pto Controlled English xi CPL CMU KANT KA VAF DAO iH 1 atural language Translation 2 ff 4 Hi A BE T Knowledge based Accurate Sen Sul A OWL Web Ontology Language ee
Download Pdf Manuals
Related Search