Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
日本語の形態素解析
Search
Yumeto Inaoka
January 26, 2017
Technology
0
430
日本語の形態素解析
2017/01/26のB3ゼミで発表
Yumeto Inaoka
January 26, 2017
Tweet
Share
More Decks by Yumeto Inaoka
See All by Yumeto Inaoka
文献紹介: Quantity doesn’t buy quality syntax with neural language models
yumeto
1
140
文献紹介: Open Domain Web Keyphrase Extraction Beyond Language Modeling
yumeto
0
180
文献紹介: Self-Supervised_Neural_Machine_Translation
yumeto
0
130
文献紹介: Comparing and Developing Tools to Measure the Readability of Domain-Specific Texts
yumeto
0
130
文献紹介: PAWS: Paraphrase Adversaries from Word Scrambling
yumeto
0
110
文献紹介: Beyond BLEU: Training Neural Machine Translation with Semantic Similarity
yumeto
0
230
文献紹介: EditNTS: An Neural Programmer-Interpreter Model for Sentence Simplification through Explicit Editing
yumeto
0
290
文献紹介: Decomposable Neural Paraphrase Generation
yumeto
0
190
文献紹介: Analyzing the Limitations of Cross-lingual Word Embedding Mappings
yumeto
0
190
Other Decks in Technology
See All in Technology
サイバー攻撃を想定したセキュリティガイドライン 策定とASM及びCNAPPの活用方法
syoshie
3
1.7k
機械学習を「社会実装」するということ 2025年版 / Social Implementation of Machine Learning 2025 Version
moepy_stats
3
160
開発生産性向上! 育成を「改善」と捉えるエンジニア育成戦略
shoota
2
830
.NET 9 のパフォーマンス改善
nenonaninu
0
2.2k
怖くない!ゼロから始めるPHPソースコードコンパイル入門
colopl
0
240
20241218_マルチアカウント環境におけるIAM_Access_Analyzerによる権限管理.pdf
nrinetcom
PRO
3
150
ネットワーク可視化の世界
likr
7
5.7k
20240522 - 躍遷創作理念 @ PicCollage Workshop
dpys
0
310
I could be Wrong!! - Learning from Agile Experts
kawaguti
PRO
8
2.5k
動画配信の フロントエンドを支える 4年間とこれから
nisshii0313
0
110
AIエージェントに脈アリかどうかを分析させてみた
sonoda_mj
2
130
【令和最新版】ロボットシミュレータ Genesis x ROS 2で始める快適AIロボット開発
hakuturu583
2
1.4k
Featured
See All Featured
XXLCSS - How to scale CSS and keep your sanity
sugarenia
248
1.3M
Stop Working from a Prison Cell
hatefulcrawdad
267
20k
[RailsConf 2023] Rails as a piece of cake
palkan
53
5.1k
Embracing the Ebb and Flow
colly
84
4.5k
BBQ
matthewcrist
85
9.4k
Design and Strategy: How to Deal with People Who Don’t "Get" Design
morganepeng
127
18k
Designing Experiences People Love
moore
139
23k
YesSQL, Process and Tooling at Scale
rocio
170
14k
Rebuilding a faster, lazier Slack
samanthasiow
79
8.8k
Writing Fast Ruby
sferik
628
61k
The World Runs on Bad Software
bkeepers
PRO
66
11k
Mobile First: as difficult as doing things right
swwweet
222
9k
Transcript
ຊޠͷܗଶૉղੳ #θϛ ࣗવݴޠॲཧݚڀࣨɹֶ෦ɹҴԬເਓ
ࢀߟʹͨ͠ॻ੶
ൃද༰ ˗ ܗଶૉղੳͰߦ͏͜ͱ ˗ ୯ޠࣙॻɾ࿈Մೳੑߦྻ ˗ ܗଶૉղੳީิͷྻڍ ˗ ղੳީิ͔Βղͷܾఆ ˗
ະޠͷॲཧ
ܗଶૉղੳ ˗ จΛ୯ޠຖʹ۠Δ ˗ ࢺΛೝࣝ͢Δ ˗ ׆༻Λೝࣝ͢Δ
ܗଶૉղੳ ˗ จΛ୯ޠຖʹ۠Δ ˗ ࢺΛೝࣝ͢Δ ˗ ׆༻Λೝࣝ͢Δ ͓ۚͰങ͑ͳ͍Ձ͕͋Δɻ
ܗଶૉղੳ ˗ จΛ୯ޠຖʹ۠Δ ˗ ࢺΛೝࣝ͢Δ ˗ ׆༻Λೝࣝ͢Δ ͓ ۚ
Ͱ ങ͑ ͳ͍ Ձ ͕ ͋Δ ɻ
ܗଶૉղੳ ˗ จΛ୯ޠຖʹ۠Δ ˗ ࢺΛೝࣝ͢Δ ˗ ׆༻Λೝࣝ͢Δ ͓ ಄ࣙ
ۚ ໊ࢺ Ͱ ॿࢺ ങ͑ ಈࢺ ͳ͍ ॿಈࢺ Ձ ໊ࢺ ͕ ॿࢺ ͋Δ ಈࢺ ɻ ۟
ܗଶૉղੳ ˗ จΛ୯ޠຖʹ۠Δ ˗ ࢺΛೝࣝ͢Δ ˗ ׆༻Λೝࣝ͢Δ ͓ ಄ࣙ
ۚ ໊ࢺ Ͱ ॿࢺ ങ͑ ಈࢺ ԼҰஈະવܗ ͳ͍ ॿಈࢺ ࿈ମܗ Ձ ໊ࢺ ͕ ॿࢺ ͋Δ ಈࢺ ޒஈऴࢭܗ ɻ ۟
୯ޠࣙॻ ୯ޠͷදهɺࢺɺ׆༻ͳͲΛ هड़ͨࣙ͠ॻ දه ࢺ ׆༻ ݩؾ ໊ࢺ ʔ
୫ ໊ࢺ ʔ ͳΔ ಈࢺ ޒஈ ʹ ॿࢺ ʔ ୲͏ ಈࢺ ޒஈ ࠜ ໊ࢺ ʔ ৸Δ ಈࢺ ԼҰஈ Β ඌࣙ ʔ
࿈Մೳੑߦྻ ͲͷΑ͏ͳࢺͷ୯ޠ͕จதͰ࿈ଓ͠ಘΔ͔Λද͢ จ ໊ࢺ ಈࢺ ॿࢺ ඌࣙ จ಄
໊ࢺ ಈࢺ ॿࢺ ඌࣙ
࿈Մೳੑߦྻ ͲͷΑ͏ͳࢺͷ୯ޠ͕จதͰ࿈ଓ͠ಘΔ͔Λද͢ จ ໊ࢺ ಈࢺ ॿࢺ ඌࣙ จ಄
໊ࢺ ಈࢺ ॿࢺ ඌࣙ
࿈Մೳੑߦྻ ͲͷΑ͏ͳࢺͷ୯ޠ͕จதͰ࿈ଓ͠ಘΔ͔Λද͢ จ ໊ࢺ ಈࢺ ॿࢺ ඌࣙ จ಄
໊ࢺ ಈࢺ ॿࢺ ඌࣙ
ܗଶૉղੳީิͷྻڍ จ಄ ͶͨΒݩؾʹͳͬͨ
ܗଶૉղੳީิͷྻڍ จ಄ Ͷ ࠜ ໊ࢺ Ͷͨ ৸Δ
ಈࢺ ͶͨΒ ৸Δ ಈࢺ ͶͨΒݩؾʹͳͬͨ
ܗଶૉղੳީิͷྻڍ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ Ͷͨ ৸Δ ಈࢺ ͶͨΒ ৸Δ ಈࢺ ͶͨΒݩؾʹͳͬͨ
ܗଶૉղੳީิͷྻڍ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ͶͨΒݩؾʹͳͬͨ
ܗଶૉղੳީิͷྻڍ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ͶͨΒݩؾʹͳͬͨ º จ ໊ࢺ ಈࢺ ॿࢺ ඌࣙ จ಄ ໊ࢺ ಈࢺ ॿࢺ ඌࣙ
ܗଶૉղੳީิͷྻڍ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ͶͨΒݩؾʹͳͬͨ º
ܗଶૉղੳީิͷྻڍ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ͶͨΒݩؾʹͳͬͨ º
ܗଶૉղੳީิͷྻڍ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ʹ ॿࢺ ͳͬͨ ͳΔ ಈࢺ ʹͳͬͨ ୲͏ ಈࢺ จ º ͶͨΒݩؾʹͳͬͨ ϥςΟεදݱ
ܗଶૉղੳީิͷྻڍ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ʹ ॿࢺ ͳͬͨ ͳΔ ಈࢺ ʹͳͬͨ ୲͏ ಈࢺ จ º ͶͨΒݩؾʹͳͬͨ
ܗଶૉղੳީิͷྻڍ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ʹ ॿࢺ ͳͬͨ ͳΔ ಈࢺ ʹͳͬͨ ୲͏ ಈࢺ จ º ͶͨΒݩؾʹͳͬͨ
ܗଶૉղੳީิͷྻڍ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ʹ ॿࢺ ͳͬͨ ͳΔ ಈࢺ ʹͳͬͨ ୲͏ ಈࢺ จ º ͶͨΒݩؾʹͳͬͨ
ܗଶૉղੳީิͷྻڍ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ʹ ॿࢺ ͳͬͨ ͳΔ ಈࢺ ʹͳͬͨ ୲͏ ಈࢺ จ º ͶͨΒݩؾʹͳͬͨ
ղੳީิ͔Βղͷܾఆ ˗ ୯ޠࣙॻɺ࿈Մೳߦྻ͚ͩͰղΛܾఆͰ͖ͳ͍ ˗ ϥςΟεߏίετΛ༩ ˗ ίετͷ࠷͍ީิΛղͱ͢Δʢίετ࠷খ๏ʣ
ϥςΟεߏͷίετͷ༩ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ʹ ॿࢺ ͳͬͨ ͳΔ ಈࢺ ʹͳͬͨ ୲͏ ಈࢺ จ º
ϥςΟεߏͷίετͷ༩ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ʹ ॿࢺ ͳͬͨ ͳΔ ಈࢺ ʹͳͬͨ ୲͏ ಈࢺ จ º
ϥςΟεߏͷίετͷ༩ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ʹ ॿࢺ ͳͬͨ ͳΔ ಈࢺ ʹͳͬͨ ୲͏ ಈࢺ จ º
֤ީิͷίετͷࢉग़ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ʹ ॿࢺ ͳͬͨ ͳΔ ಈࢺ ʹͳͬͨ ୲͏ ಈࢺ จ º
֤ީิͷίετͷࢉग़ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ʹ ॿࢺ ͳͬͨ ͳΔ ಈࢺ ʹͳͬͨ ୲͏ ಈࢺ จ º
֤ީิͷίετͷࢉग़ ˗ ࠜɹ୫ɹݩؾɹʹɹͳͬͨ ˗ ࠜɹ୫ɹݩؾɹ୲ͬͨ ˗ ৸ͨΒɹݩؾɹʹɹͳͬͨ ˗ ৸ͨΒɹݩؾɹ୲ͬͨ
ˠ ˠ ˠ ࠷খ ˠ
࠷খίετͷީิͷ୳ࡧ ˗ ޠኮͷ૿ՃɺೖྗจͷจԽͰީิരൃతʹ૿Ճ ˠશީิͷίετࢉग़࣮࣭ෆՄೳ ˗ ೖྗจΛׂ͠ɺͦΕͧΕͷ݁ՌΛ౷߹ ˠೖྗจͷจԽʹΑΔީิͷ૿ՃΛ͑ΒΕΔ
ϏλϏΞϧΰϦζϜ ˗ ಈతܭը๏ %ZOBNJD1SPHSBNNJOH%1 ͷͻͱͭ ˗ .F$BCɺ$IBTFOͰ࠾༻͞Ε͍ͯΔ ˗ ӳจͷࢺղੳͳͲʹ༻͞Ε͍ͯΔ
ϏλϏΞϧΰϦζϜ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ʹ ॿࢺ ͳͬͨ ͳΔ ಈࢺ ʹͳͬͨ ୲͏ ಈࢺ จ º
ϏλϏΞϧΰϦζϜ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ʹ ॿࢺ ͳͬͨ ͳΔ ಈࢺ ʹͳͬͨ ୲͏ ಈࢺ จ º
ϏλϏΞϧΰϦζϜ จ಄ Ͷ ࠜ ໊ࢺ ͨΒ ୫
໊ࢺ ݩؾ ໊ࢺ Ͷͨ ৸Δ ಈࢺ Β ඌࣙ ͶͨΒ ৸Δ ಈࢺ ʹ ॿࢺ ͳͬͨ ͳΔ ಈࢺ ʹͳͬͨ ୲͏ ಈࢺ จ º
ະޠͷॲཧ ˗ จதʹະޠ͕ग़ΔͱϥςΟεߏ͕࡞ΒΕͳ͍ ˠղੳΛߦ͏͜ͱ͕Ͱ͖ͳ͍ ˗ ࣈɾΧλΧφͷ࿈ଓΛҰޠͱ͠ɺߴ͍ίετΛ༩ ˠࣙॻͷޠͷΈͰղऍͰ͖ΔͳΒͦͪΒΛ༻ ˠແཧ͕͋ΔͳΒٖࣅతͳޠΛ༻
ൃදͨ͠༰ ˗ ܗଶૉղੳͰߦ͏͜ͱ ˗ ୯ޠࣙॻɾ࿈Մೳੑߦྻ ˗ ܗଶૉղੳީิͷྻڍ ˗ ղੳީิ͔Βղͷܾఆ ˗
ະޠͷॲཧ