Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Sign up for free
Menu
Search
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Pricing
Search
Sign in
Sign up for free
ATPツアーレベルにおけるサーブ制球力の分析
Search
AkinohiAkatsuki
May 30, 2020
230
0
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
ATPツアーレベルにおけるサーブ制球力の分析
AkinohiAkatsuki
May 30, 2020
More Decks by AkinohiAkatsuki
See All by AkinohiAkatsuki
サーブの制球力を定量的に評価してみよう(テニスのはなし)
akinohiakatsuki
0
1.1k
Featured
See All Featured
Build your cross-platform service in a week with App Engine
jlugia
234
19k
WCS-LA-2024
lcolladotor
0
890
コードの90%をAIが書く世界で何が待っているのか / What awaits us in a world where 90% of the code is written by AI
rkaga
63
46k
Beyond borders and beyond the search box: How to win the global "messy middle" with AI-driven SEO
davidcarrasco
3
280
ピンチをチャンスに:未来をつくるプロダクトロードマップ #pmconf2020
aki_iinuma
128
56k
How to train your dragon (web standard)
notwaldorf
96
6.9k
StorybookのUI Testing Handbookを読んだ
zakiyama
31
7k
Marketing Yourself as an Engineer | Alaka | Gurzu
gurzu
0
330
Odyssey Design
rkendrick25
PRO
2
880
What does AI have to do with Human Rights?
axbom
PRO
1
2.5k
Darren the Foodie - Storyboard
khoart
PRO
4
4k
Google's AI Overviews - The New Search
badams
0
1.6k
Transcript
"51πΞʔϨϕϧʹ͓͚Δ αʔϒ੍ٿྗͷੳ
ੳαϚϦʔ എܠ ༻σʔλ ੳ݁Ռ • αʔϒͷʮίϯτϩʔϧ͕ྑ͍ʯͬͯͲ͏ ͍͏͜ͱʁ • αʔϒͰૂ͏͖Ͳ͜ʁ
• ຊʹίϯτϩʔϧ͕ྑ͍ͷ୭͔ʁ "51ެࣜαΠτʹܝࡌ͞Ε͍ͯΔσʔλ • ணͷ Y Z ࠲ඪ • ϙΠϯτಘࣦͷ݁Ռ • ίʔτͷঢ়گ • બखͷར͖खɾ ͳͲ • ࠨӈ྆ʹ͍ۙαʔϒ͕ྑ͍ɻਂ͋͞· Γҙຯͳ͠ • ʹΑͬͯૂ͏͖κʔϯ͕ҟͳΔ • ணͱϥΠϯͱͷڑʹ͋·Γҙຯͳ ͍ • ʢӈWTӈͷͱ͖ʣ૬खӈଆͷ΄͏͕֫ಘ ߴ͍ TUαʔϒ OEαʔϒ
ੳͷഎܠ Ұൠͱͯ͠ɺϥΠϯࡍʹूΊΒΕ͍ͯΔ΄͏͕ྑ͍ͱ͞ΕΔ αʔϒͷʮίϯτϩʔϧ͕ྑ͍ʯͱ Ͳ͏͍͏͜ͱ͔ʁ
ੳͷഎܠ 2ͳΒɺϥΠϯࡍͷʮࡍʯͱͲͷ͘Β͍ͷڑʁ 2ͲͷϥΠϯ͔Βͷڑಉ͡ॏΈΛ࣋ͭͷʁ 2ྑ͍ʗѱ͍ͷڥͲ͜ʹ͋Δʁ 2બखؒͰ੍ٿྗʹ༏ྼΛ͚ͭΔͱͨ͠ΒͲ͏ͬͯʁ
ੳͷഎܠ 57தܧͰ࣌ʑɺٿ͕ՄࢹԽ͞ΕΔ͕ৄࡉෆ໌ ํͷͲͷϥΠϯ͔ΒͷڑϙΠϯτΛ֫ಘ͢Δ͏͑ͰՁͳͷ͔ʁ ը૾ग़ॴɿ5FOOJT57 ΦϨϯδͷҹͱΞϧϑΝϕοτൃදऀʹΑΔ ྫ͑͜ͷը૾தͷ"ͱ#͕ ࣋ͭҙຯຊʹಉ͔͡ʁ " #
༻σʔλհ ࢼ߹͝ͱͷணσʔλΛ4FMFOJVNͰεΫϨΠϐϯά ˠ σʔλϑϨʔϜԽ ը૾ͱͯ͠ݟΔͨΊͷͷ͕ͩ47(λάͳͷͰ࠲ඪ͕ଘࡏ͢Δ બखͷར͖ खϓϩϑΟʔϧ ϖʔδʹඈΜͰऔ ಘ͠ɺඥ͚ͮͨ
༻σʔλհ σʔλͷ͋Δͷશࢼ߹ɺສ ٿͷαʔϒΛͱʹੳ
༻σʔλհ औಘͨ͠σʔλҎԼͷͱ͓Γ ໊শ ༰ 4FSWFS αʔϒΛଧͬͨબखͷ໊લ 3FUVSOFS αʔϒΛฦ͢ଆͷબखͷ໊લ Y
ணͷ9࠲ඪ Z ணͷ:࠲ඪ 4FDUJPO 9࠲ඪͷ۠ 1PJOUT8@- ϙΠϯτͷಘࣦ "DF αʔϏεΤʔε͔ 4FSWF TUαʔϒ͔ɺOEαʔϒ͔ $PVSU %FVDFଆ͔ɺ"Eଆ͔ ໊শ ༰ )FJHIU ') ӈଧ͔ͪɺࠨଧ͔ͪ #) ྆खଧ͔ͪɺยखଧ͔ͪ .BUDI8JOOFS ࢼ߹ͷউऀ 5PVSOBNFOU େձ໊ 3PVOE Կճઓ͔ 4VSGBDF ίʔτͷࡐ࣭ 7FOVF ͔ɺ֎͔ 4FBTPO Կͷγʔζϯ͔ * ͜ͷεϥΠυ্Ͱͻͱ·ͱΊʹ͍ͯ͠Δ͕ɺ࣮ࡍ 4FSWFS ͱ 3FUVSOFS ͰͦΕͧΕผʹऔಘ͠ྻΛ͚͍ͯΔ ଠࣈࠓճͷੳͰओʹ༻ͨ͠ͷ
ੳͷఔ લॲཧ l ޡ৹ɺޡܭଌͱࢥΘΕΔσʔλΛআ֎ l ʮӈଧͪ WT ӈଧͪʯ͔ͭʮϋʔυίʔτʯʹ
ߜΓࠐΉ l αʔόʔͷΛDNࠁΈͰάϧʔϓԽ ՄࢹԽ l TUαʔϒɺOEαʔϒ͝ͱʹணΛϓϩοτ ͠ɺٿͷΛ၆ᛌ l ώετάϥϜͰ9࠲ඪɺ:࠲ඪͱϙΠϯτಘࣦͷ ؔΛѲ ૂ͏͖κʔϯͷಋग़ l αϙʔτϕΫλʔϚγϯͰίʔτฏ໘্ͷ࠲ඪ ͔ΒϙΠϯτͷಘࣦΛ༧ଌ l ܾఆڥքΛϓϩοτ దਖ਼ͳٿόϥϯεͷಋग़ l ࢼ߹͝ͱͷಛఆηΫγϣϯͷٿʢภΓͷ ڧ͞ʣͱɺϙΠϯτ֫ಘͷ૬ؔΛݟΔ
ิʛ༻ޠղઆ ը૾ग़ॴɿIUUQXXXUBDUJDBMUFOOJTDPNBMFYEFNJOBVSTNBSUFTULJEDMBTTSPPN αʔόʔ Ϧ λ ồ φ ồ %FVDFίʔτ
"Eίʔτ αʔϏεϥΠϯ ηϯλʔϥΠϯ αΠυϥΠϯ αΠυϥΠϯ αʔϒΛ%FVDFίʔτʹ ଧͭͱ͖͜ͷൣғ͔Β αʔϒΛ"Eίʔτʹଧͭͱ͖ ͜ͷൣғ͔Β Ξυ σϡʔε
ิʛ༻ޠղઆ ը૾ग़ॴɿ"51 8JEFɿԫ৭ ίʔτ྆ͷηΫγϣϯɻʮϫΠυʯ Ҏ֎ͷݺͼํී௨͠ͳ͍ɻ .JEɿ ϫΠυͱ5ͷதؒͳͷͰ.JEɻ͜͜ʹ ணͨ͠αʔϒ૬खͷମʹ͔ͬͯ ͍͘ͷͰʮϘσΟαʔϒʯͱ͍͏ɻ
5ɿ αʔόʔଆ͔ΒݟΔͱαʔϏεϥΠϯ ͱηϯλʔϥΠϯ͕ʮ5ʯͷࣈΛඳ͍ ͍ͯΔͷͰ5ɻผ໊ʮηϯλʔʯɻ ωοτʹ͍ۙ ΄Ͳʮઙ͍ʯ ͱදݱ͢Δ αʔϏεϥΠ ϯʹ͍ۙ΄Ͳ ʮਂ͍ʯͱද ݱ͢Δ
ิʛ༻ޠղઆ TUαʔϒ l ϙΠϯτ͝ͱαʔϒΛଧͭνϟϯε͕ճ͋Γɺͦͷ͏ͪͷճ l ೖΕͦͷ··ϓϨʔଓߦ l ֎ͯ͠ճͷνϟϯεʢOEαʔϒʣ͕͋ΔͷͰࡍͲ͍ͱ͜Ζڧ͘ଧͭ OEαʔϒ
l TUαʔϒ͕֎Εͨͱ͖ʹଧͭ l ೖΕϓϨʔଓߦ l ͜Ε֎͢ͱμϒϧϑΥϧτͱͳͬͯ૬खͷಘʹͳΔ l ֎͢ͱ·͍ͣͷͰεϐʔυΛམͱͯ͠৻ॏʹೖΕΔ
લॲཧ ҟৗΛআ֎ ϥΠϯ͔Β֎Ε͍ͯΔʹ͔͔ΘΒͣʮೖ͍ͬͯΔʯͱఆ͞Ε͍ͯΔαʔϒΛআ֎ɻ ϋʔυίʔτʹݶఆ ίʔτͷࡐ࣭ʹΑΓϘʔϧͷΈํɺݮͷํɺͷऔΒΕํ͕େ͖͘ҟͳΔͨΊɺ࠷ࢼ߹ͷଟ ͍ϋʔυίʔτʹݶͬͨɻ ϋʔυίʔτͱɺίϯΫϦʔτΞεϑΝϧτͷجૅͷ্ʹथࢷͷίʔςΟϯάΛͨ͠ͷɻ ӈଧͪWTӈଧͪʹݶఆ ͷࠨӈͰαʔϒͷճసํ͕స͢ΔͨΊɺணʹ͕ࠩͳͯ͘ੑ࣭໌Β͔ʹҟͳΔɻ
Ϧλʔφʔ৳͕͢ٯʹͳΔ͜ͱ͔ΒɺଧͨΕͨαʔϒͷண͕ಉҰͰରॲͷํҟͳΔɻ
લॲཧ αʔόʔͷΛάϧʔϓ͚ Nͷબख͕·͙ͬ͢LNIͰଧͬͨαʔϒ ͱɺDNͷબख͕LNIͰଧͬͨαʔϒ ɺண͕ಉҰͰ໌Β͔ʹಉ࣭Ͱͳ͍ɻ αʔϒωοτΛӽ͑ɺ͔ͭαʔϏεϥΠϯҎ ʹೲΊͳ͍͚ͯ͘ͳ͍ͨΊɺҰൠʹഎ͕ ͍બख΄Ͳ͘ɾճసΛ͔͚Δඞཁ͕͋Δɻ ࠲ඪͷม
ʮϥΠϯࡍʯͷڑΛܭࢉ͘͢͢͠ΔͨΊɺ 9࠲ඪɾ:࠲ඪΛʮ֤ϥΠϯ͔Βͷڑʯ͓Α ͼʮ࠷دΓͷϥΠϯ͔Βͷڑʯʹࢉͨ͠ɻ ࣮ࡍʹ༻ͨ͠ͷສ ٿͷαʔϒ
ՄࢹԽ
ิʛࢄਤͷݟํ ը૾ग़ॴɿ5FOOJT57 Ҏ߱ͷϖʔδͰڍ͛ΔࢄਤɺಛʹஅΓ͕ͳ͍ݶΓԼهΦϨϯδͷκʔϯ Λ্͔Βݟͨͱ͖ͷணΛද͍ͯ͠Δɻ
ՄࢹԽ • TUαʔϒ%FVDF "Eͱʹࠨӈ྆ʹ͍ۙ΄ͲϙΠϯτ͕औΓ͍͢ • OEͲ͜ʹଧͬͯಉ͡Α͏ͳ֫ಘʹݟ͑Δ ʢฦ͞Ε͍͢ͷͰαʔϒͷ༏Ґੑ͕͘ɺ΄΅ϥϦʔͷग़དྷͰϙΠϯτ͕ܾ·Δʣ • OE#)ʢӈଧͪͷબखͳΒࠨʣଆΛ߈ΊΔηΦϦʔʹଇͬͨٿ͕ݟͯͱΕΔ
TUαʔϒͷ9࠲ඪͱϙΠϯτಘࣦͷώετάϥϜ "%ίʔτ %FVDFίʔτ ࠨӈ྆ʹ͍ۙ΄Ͳ֫ಘ͕ߴ͍ɻηϯλʔΑ ΓϫΠυͷ΄͏͕༗ޮʹݟ͑Δɻ ࠨӈ྆ʹ͍ۙ΄Ͳ֫ಘ͕ߴ͍ɻϫΠυͱη ϯλʔʹҧ͍ಛʹݟͯͱΕͳ͍ɻ
TUαʔϒͷ:࠲ඪͱϙΠϯτಘࣦͷώετάϥϜ "%ίʔτ %FVDFίʔτ ਂ͍΄Ͳྑ͍αʔϒͱݴ͍͕͍ͨɻͨͩ͠ɺ ઙ͗͢Δͱ֫ಘ͕͍ɻ "Eίʔτͱಉ༷ͷɻ
OEαʔϒͷ9࠲ඪͱϙΠϯτಘࣦͷώετάϥϜ Ͳ͜ʹଧͬͯ΄΅Ұఆͷ֫ಘɻηϯλʔ͕ ༗ޮɻ "Eίʔτͱಉ༷ͷɻϫΠυ͕༗ޮɻ "%ίʔτ %FVDFίʔτ
OEαʔϒͷ:࠲ඪͱϙΠϯτಘࣦͷώετάϥϜ ਂ͍΄Ͳྑ͍αʔϒͱݴ͍͕͍ͨɻNΛ ͑ͯઙ͍ͱ֫ಘ͕͍ɻ NΛ͑ͯઙ͍ͱ֫ಘ͕͍͕ɺਂ͍΄ Ͳྑ͍αʔϒͱݴ͑ͳ͍ɻ "%ίʔτ %FVDFίʔτ
ՄࢹԽ લड़ͷ"ͱൺͯ#΄ͱΜͲҙຯΛ࣋ͨͳ͍ ը૾ग़ॴɿ5FOOJT57 ΦϨϯδͷҹͱΞϧϑΝϕοτൃදऀʹΑΔ " #
ՄࢹԽ TUαʔϒͱOEαʔϒͷάϧʔϓผαʔϒண • TUαʔϒ͕ߴ͘ͳΔ΄Ͳઙ͍Ґஔʹ͕͍ͬͯΔ • OEαʔϒ͝ͱͷҧ͍͕ಡΈऔΓʹ͍͘ • DNDN͘͝গʢͦͦੈքϥϯΩϯά501ʹ໊͔͍͠ͳ͍ʣ
ίϯτϩʔϧ͕ྑ͍ͱݴ͑Δκʔϯͷಋग़
ՄࢹԽ͔ΒͷࣔࠦͱରԠ ՄࢹԽ͔Βͷࣔࠦ • OEαʔϒ࠲ඪ͔Β֫ಘΛ༧ଌ͢Δ͜ͱ͕ۃΊͯࠔ • αʔόʔͷʹΑͬͯૂ͏ʢૂ͑Δʣൣғ͕ҟͳΔ • DNɾDNʹภ͍ͬͯΔ ରԠࡦ
• ༗ޮͳκʔϯTUαʔϒʹݶͬͯɺάϧʔϓผʹಋ͘ • DNͱDNΛલͱޙʹͦΕͧΕׂ • DNDNҎ্ͱͯ͠ͻͱ·ͱΊʹ
ϞσϦϯά جຊతͳߟ͑ํ ਫ਼ΑΓղऍՄೳੑΛॏࢹ ˠࢼ߹தࢼ߹લޙʹʮࠓͷαʔϒྑ͍ͱ͜ΖʹίϯτϩʔϧͰ͖͍ͯΔʗ͍ͳ͍ʯ ͱධՁ͢ΔஅࡐྉΛ࡞Γ͍ͨͨΊɺܾఆڥքෳࡶʹͳΒͳ͍Α͏ʹͨ͠ ༻͢Δಛྔ • 9࠲ඪͱ:࠲ඪͷม ʢٿɾճసྔɾճసํͷσʔλެ։͞Ε͍ͯͳ͍ʣ
• తม 1PJOUT8PO-PTUͷʢྨʣ ˠ 8POΛɺ-PTUΛͱͨ͠
ΞϧΰϦζϜͷબ ϥϯμϜϑΥϨετͱαϙʔτϕΫλʔϚγϯͷ྆ํΛࢼͯ͠ൺֱ ˠ αϙʔτϕΫλʔϚγϯΛ࠾༻ ཧ༝ɿ ਫ਼΄΅ಉ͕ͩɺϥϯμϜϑΥϨετٿͷಛΛ͋·Γଊ͍͑ͯͳ͍ͨΊɻαʔϒࣼΊ ํʹଧͭͨΊʮീʯͷࣈঢ়ʹ͢Δ͕ɺ8POͱ༧ଌ͞ΕΔൣғΛϓϩοτ͢Δͱίʔτͷࠨӈ ʹॎͷଳ͕Ͱ͖͍ͯΔ͚ͩɻ ཧ༝ɿ
ϥϯμϜϑΥϨετͷ΄͏͕ૂ͍Α͏ͷͳ͍ൣғΛࣔͯ͠͠·͍͕ͪͳͨΊɻաֶशΛආ͚ΔͨΊ ʹύϥϝʔλʔΛௐͯ͠ɺ෯DNఔͷࡉ͍ଳ͕ਫฏʹΔͳͲɺʮ୭͕ͦ͜Λૂͬͯίϯτ ϩʔϧͰ͖Δͷ͔ʁʯͱ͍͏ඇݱ࣮తͳ༧ଌΛฦ͖ͯͨ͠ɻ
ύϥϝʔλʔͷௐ ෆۉߧͷௐ TUαʔϒͷ֫ಘฏۉ͢ΔͱఔͳͷͰ 8POɿ-PTU͓͓ΉͶɿɻΞϯμʔαϯϓϦ ϯάΛߦ͏΄ͲͷภΓͰͳ͍ͱߟ͑ɺ class_weight='balanced' ͰରԠɻ աֶशͷ੍ جຊతͳߟ͑ํʹԊ͏ͨΊɺ
C ͱ gamma ඞͣ͠ྨਫ਼Λٻͤͣʹௐͨ͠ɻ ਫ਼ͱ൚Խੑೳͷ্ ׂަࠩݕূͱάϦουαʔνΛ࣮ࢪɻάϦουαʔνͷࡍɺ෯Λ࣋ͬͯαʔνˠग़ྗ݁ՌΛ ͱʹൣғΛߜͬͯαʔνʜʜΛճ܁Γฦͨ͠ɻධՁࢦඪ roc_auc Λબɻ
ڥքܾఆઢɿDNະຬͷ߹ "%ίʔτ %FVDFίʔτ accuracy : 0.4730 f-measure : 0.5021
precision : 0.7406 AUC : 0.5349 recall : 0.3798 accuracy : 0.5494 f-measure : 0.6078 precision : 0.7795 AUC : 0.5839 recall : 0.4981
ڥքܾఆઢɿDNҎ্DNະຬͷ߹ "%ίʔτ %FVDFίʔτ accuracy : 0.5509 f-measure : 0.6054
precision : 0.7881 AUC : 0.5909 recall : 0.4914 accuracy : 0.5398 f-measure : 0.5701 precision : 0.8246 AUC : 0.6094 recall : 0.4357
ڥքܾఆઢɿDNҎ্DNະຬͷ߹ "%ίʔτ %FVDFίʔτ accuracy : 0.5426 f-measure : 0.5859
precision : 0.8134 AUC : 0.6023 recall : 0.4578 accuracy : 0.5684 f-measure : 0.6248 precision : 0.8401 AUC : 0.6254 recall : 0.4973
ڥքܾఆઢɿDNҎ্DNະຬͷ߹ "%ίʔτ %FVDFίʔτ accuracy : 0.5562 f-measure : 0.6236
precision : 0.8401 AUC : 0.6126 recall : 0.4958 accuracy : 0.5847 f-measure : 0.6679 precision : 0.8441 AUC : 0.6185 recall : 0.5525
ڥքܾఆઢɿDNҎ্DNະຬͷ߹ "%ίʔτ %FVDFίʔτ accuracy : 0.5754 f-measure : 0.6531
precision : 0.8509 AUC : 0.6224 recall : 0.5300 accuracy : 0.5928 f-measure : 0.6632 precision : 0.8482 AUC : 0.6362 recall : 0.5444
ڥքܾఆઢɿDNҎ্ͷ߹ "%ίʔτ %FVDFίʔτ accuracy : 0.6331 f-measure : 0.7213
precision : 0.9167 AUC : 0.6902 recall : 0.5946 accuracy : 0.6409 f-measure : 0.7290 precision : 0.8764 AUC : 0.6613 recall : 0.6240
ڥքܾఆઢɿάϧʔϓผҰཡ DN DN DN DN DN DN %FVDF "E
͕ߴ͍΄ͲɺϫΠυͷઙ͍ҐஔͰ੨͍κʔϯ͕ΒΜͰ͍Δ ˠଧߴ͕ߴ͍ɺೖࣹ֯Λখͯ͘͞͠खલʹଧͯΔ͜ͱ͕ө ͕ߴ͍΄Ͳɺྨਫ਼͕ߴ͍ ˠ ͕͍΄Ͳαʔϒ͕͘ϥϦʔʹͳΓ͍ͨ͢Ίɺෆ֬ఆཁૉ͕ଟ͍͜ͱ͕ө
ڥքܾఆઢɿͰ͚ͳ͍߹ "%ίʔτ %FVDFίʔτ accuracy : 0.5540 f-measure : 0.6074
precision : 0.8177 AUC : 0.6071 recall : 0.4832 accuracy : 0.5638 f-measure : 0.6200 precision : 0.8507 AUC : 0.6284 recall : 0.4877 ϫΠυͷ΄͏͕༗ޮκʔϯ͕͘ɺઙ ͍΄͏͕ޮՌతɻ ηϯλʔͱϫΠυͷκʔϯͷ͞ۉ ʹ͍ۙɻηϯλʔਂ͍΄ͲɺϫΠ υઙ͍΄Ͳྑ͍ɻ
༗ޮͳκʔϯͳͥࣼΊʁ αʔϒࣼΊํʹଧͭ ˠ ಛʹϫΠυઙ͍΄ͲϦλʔφʔ͔Βԕ͍ ઙ͍ϫΠυαʔϒࠨӈ͕ଆʹೖͬͯɺਂ͍߹ͱࣅͨޮՌΛ࣋ͭ
༗ޮͳκʔϯͳͥࣼΊʁ
ੳ݁Ռͷ׆༻ ࢼ߹தࢼ߹ޙͷৼΓฦΓ࣌ʹɺ࣮ࡍͷணͱ੨͍κʔϯΛরΒ͠߹Θͤ Δ͜ͱͰίϯτϩʔϧ͕ྑ͔ͬͨͷ͔ɺѱ͔ͬͨͷ͔͕ఆྔతʹΘ͔Δɻ • ʦྫʧ੨͍κʔϯೖׂͬͨ߹Λൺֱ͠ɺͲͪΒͷબख͕ΑΓίϯτϩʔϧʹ༏Ε͍ͯ Δ͔͕ࣈͱͯ͠ൺֱͰ͖Δɻ • ʦྫʧࢼ߹Λ௨ͯͯ͢͡ͷαʔϒ͕੨͍κʔϯʹೖΓͳ͕ΒϙΠϯτ͕͋·ΓऔΕ ͳ͍߹ɺαʔϒͷૂ͍͕͍͍ͤͰͳ͍ͱݴ͑Δɻ
• ʦྫʧ੨͍κʔϯΛ֎ΕΔαʔϒ͕ଟ͔ͬͨʹ͔͔ΘΒͣϙΠϯτ͕औΕΔ߹ɺ૬ खͷϦλʔϯ͕ྑ͘ͳ͍ͱΘ͔Δɻ
ͪͳΈʹɺϫΠυͱηϯλʔɺͲͪΒଆΛૂ͏͖͔
ੳ͔Βͷࣔࠦͱٙ ૂ͏͖ൣғΘ͔͕ͬͨɺϫΠυͱηϯλʔͷͲͪΒଆΛૂ͑ྑ͍͔ʁ • αʔόʔ͕ϫΠυ͔͠ૂΘͳ͍ͳΒɺϦλʔφʔηϯλʔΛܯռ͢Δඞཁ͕ͳ͍ • αʔόʔ͕ηϯλʔ͔͠ૂΘͳ͍ͳΒɺϦλʔφʔϫΠυΛܯռ͢Δඞཁ͕ͳ͍ • ޒޒʹଧ͚ͪΔͷ͕ྑ͍ͷ͔ɺɿͷΑ͏ʹଟগภΒͤΔ΄͏͕ྑ͍ͷ͔ ˠ
ࢼ߹͝ͱʹɺTUαʔϒͷϫΠυͷภΓɺηϯλʔͷภΓͱɺͦͷࢼ ߹ͰͷϙΠϯτ֫ಘΛٻΊϓϩοτ
TUαʔϒͷٿͷภΓͱϙΠϯτ֫ಘ "EίʔτͰTUαʔϒͷԿׂ ΛϫΠυଆʹଧ͔ͬͨ "EίʔτͰTUαʔϒͷԿׂ Ληϯλʔଆʹଧ͔ͬͨ %FVDFίʔτͰTUαʔϒͷ ԿׂΛηϯλʔଆʹଧ͔ͬͨ %FVDFίʔτͰTUαʔϒͷ ԿׂΛϫΠυଆʹଧ͔ͬͨ
ͯ͢ϙΠϯτ֫ಘͱͷ૬ؔະຬ ˠ ҰํʹภΒͤͯ྆ํʹଧ͚ͪͯɺࢼ߹݁Ռͱ૬͕ؔͳ͍
ੳͷҙ ࢼ߹Λ௨ͯ͡ͷٿ͔͠Έ͍ͯͳ͍ • 8ˠ5ˠ8ˠ5ˠ8ˠ5 8ˠ8ˠ8ˠ5ˠ5ˠ5ಉ͡ͷͱ͍ͯ͠Δ͕ɺ࣮ࡍʹ ʮύλʔϯʯͷӨڹΛड͚Δͣ ٿɾٿछͷҧ͍Λߟྀ͍ͯ͠ͳ͍ • ࣮ࡍͷαʔϒʹٿɺճసྔɺճసํͳͲͷҧ͍͕͋ΔͨΊɺͦͷ͍͚ϙΠ
ϯτ֫ಘʹӨڹ͢Δͣ
OEαʔϒͲ͜Λૂ͏͖͔
OEαʔϒͰૂ͏͖ॴ͋Δͷ͔ • OEαʔϒͷ8-Λ࠲ඪ͔Βྨ͢Δ͜ͱࠔ ˠ ϙΠϯτಘࣦΛண͔Β༧ଌ͢Δ͜ͱࠔ • ΘΓʹɺ#)ଆͷूதͱ͍͏ηΦϦʔʹҙຯ͕͋Δͷ͔Λੳ ˠ ૂ͏͖େ·͔ʹ')ଆͱ#)ଆͷͲͪΒ͔
OEαʔϒͷηΫγϣϯผ֫ಘͷൺֱ • ࢼ߹͝ͱͷOEαʔϒͷϙΠϯτ֫ಘΛηΫγϣϯผʹࢉग़ • #)ଆʢΦϨϯδʣʹରͯ͠')ଆʢ੨ʣͷ΄͏͕ߴ͍ ˠ ౷ܭ্࣮࣭తʹ͕ࠩ͋Δͷ͔ʁ "%ίʔτ %FVDFίʔτ
OEαʔϒͷηΫγϣϯผ֫ಘͷݕఆ • ໌Β͔ʹࢄͰͳ͘ਖ਼نੑͳ͍ͨΊɺϚϯϗΠοτχʔͷ6ݕఆΛ࣮ࢪ ʢ')#)ͱ͢Δยଆݕఆɺ༗ҙਫ४ʣ • ͍ͣΕ QͷͨΊɺ#)ଆΑΓ')ଆΛૂͬͨ΄͏͕Α͍ͱݴ͑Δ "%ίʔτ %FVDFίʔτ
MannwhitneyuResult: pvalue=1.716e-10 MannwhitneyuResult: pvalue=0.0013