前処理勉強会_発表資料_MITTI_20210724.pdf
by
mitti1210
Link
Embed
Share
Beginning
This slide
Copy link URL
Copy link URL
Copy iframe embed code
Copy iframe embed code
Copy javascript embed code
Copy javascript embed code
Share
Tweet
Share
Tweet
Slide 1
Slide 1 text
લॲཧΛRͰͬͯΈͨʂ DPCσʔλʹઓʂ 1 2021/07/24 MITTI
Slide 2
Slide 2 text
ϑΥϧμͷ४උ .*55*ͷൃදͰҎԼͷߏͰϑΥϧμΛ࡞͍ͬͯ·͢ 3ͰͰ͖·͕࣌ؒ͢ͷ߹্ল͘ͷͰࣄલʹ࡞͢ΔͱਐΊ͍͢Ͱ͢ 2 ϓϩδΣΫτ ϑΥϧμ JOQVU TVNNBSZ .%$@@ PVUQVU JOQVU .%$@@
Slide 3
Slide 3 text
μϯϩʔυ 3 ͜ͷ4 ࢪઃ֓ཁද ʢ̔ʣ࣬ױผखज़ผूܭ MDC07 ˎ2016,20172ͭʹ͔Ε͍ͯΔ MDC07ɿܗ֎Պ
Slide 4
Slide 4 text
μϯϩʔυͱอଘϑΝΠϧ໊ ͜ͷϑΥϧμ໊ͱϑΝΠϧ໊ͰਐΊ·͢ 4 ϓϩδΣΫτ ϑΥϧμ JOQVU TVNNBSZ .%$@@ PVUQVU TVNNBSZ .%$@@ ʢ̔ʣ࣬ױผखज़ผूܭ MDC07 ࢪઃ֓ཁද
Slide 5
Slide 5 text
ࣄલ४උ 5 #ࠓճtidyverseύοέʔδΛ͍·͢ɻ #Πϯετʔϧͨ͜͠ͱͳ͍ํઌʹΠϯετʔϧ install.packages(“tidyverse") #ͲͪΒͷൃදͰ͍·͢ #Πϯετʔϧͨ͠ύοέʔδͷݺͼग़͠ library(tidyverse) #ࠓճಛʹdplyr,tidyr,stringr,purrrΛ༻ library(readxl) #excelͷಡΈࠐΈʹ༻
Slide 6
Slide 6 text
ୀӃױऀௐࠪ 6 https://www.mhlw.go.jp/stf/shingi2/0000196043_00003.html
Slide 7
Slide 7 text
ख࡞ۀͰͰ͖Δʁ 7 ෳྻ໊ ࡏӃ ͍Βͳ͍ ϋΠϑϯ ͍Βͳ͍ ͕มΘΔͱ ૿͑ͨΓݮͬͨΓ͢Δ ͕มΘΔͱ පӃ໊͕มΘΔ͜ͱ͕͋Δ ͕มΘΔͱ ප໊มΘΔ͜ͱ͕͋Δ ԣྻɺॎߦ ºؒ ʢ༌݂Ҏ֎ͷ࠶ܝʣ Λআ͍͕ͨ߹ܭ ͕มΘΔͱ ൪߸ҧ͏͜ͱ͕͋Δ ௨൪ʹલͷͷࠂࣔ൪߸
Slide 8
Slide 8 text
͜ΕͳΒੳʹ͔͚ͨΓάϥϑ͕࡞ΕΔ 8 ͰมΘΔ൪߸Λ ἧ͑ͨ ࢢொଜ໊͚ͨ ෳΛͭͳ͛ͨ ප໊ͰݕࡧͰ͖Δ ਓ͚ͩʹͨ͠
Slide 9
Slide 9 text
ࠓճͷલॲཧ ˔UJEZʢવσʔλʣʹ͢Δ ˔จࣈྻॲཧ ˔܁Γฦ͠ॲཧʹରԠ͢Δ ˔ؔͰ·ͱΊΔ 9
Slide 10
Slide 10 text
UJEZʢવσʔλʣͱʁ 8JLJQFEJB༷ΑΓ IUUQTKBXJLJQFEJBPSHXJLJ5JEZ@EBUB 10
Slide 11
Slide 11 text
σʔλͷܗɿUJEZS 11 ࡶવσʔλ %>% pivot_longer ( col = c("6࣌", "12࣌", “18࣌”) , names_to = “࣌ࠁ” , values_to = “ఱؾ”)
Slide 12
Slide 12 text
UJEZʹ͢ΔͨΊͷઓ͍ 12 5JEZ /PU5JEZ
Slide 13
Slide 13 text
ࠓճͷલॲཧ ˔UJEZʢવσʔλʣʹ͢ΔɿUJEZS ˔จࣈྻॲཧɿTUSJOHS ˔܁Γฦ͠ॲཧʹରԠ͢ΔɿQVSSS ˔ؔͰ·ͱΊΔɿGVODUJPO ˔͚ͬͭ͘ΔɿKPJO 13
Slide 14
Slide 14 text
จࣈྻॲཧɿTUSJOHS 14 TUSJOHS දతͳؔ આ໌ ݕࡧ TUS@EFUFDU ͋Δจࣈྻؚ͕·Ε͍ͯͨΒ536& நग़ TUS@FYUSBDU TUS@FYUSBDU@BMM TUS@TVC ͋ΔύλʔϯʹҰக͢ΔจࣈྻΛநग़ ࠷ॳʢ࠷ޙʣ˓จࣈΛநग़ ஔ͖͑ TUS@SFQMBDF TUS@SFQMBDF@BMM ύλʔϯʹϚονͨ͠ՕॴΛஔ Ճ TUS@D จࣈྻΛ͚ͬͭ͘Δ https://kazutan.github.io/kazutanR/stringr-intro.html ͕ৄ͍͠Ͱ͢ʂ
Slide 15
Slide 15 text
TUSJOHSͱਖ਼نදݱ 15 ྫʣ“MDC_02_8_07_2019.xlsx"ͱ͍͏ϑΝΠϧ໊͔Β৭ʑऔΓग़͍ͨ͠ str_extract(“จࣈྻ”, “݅” ) ݅ʹ৭ʑͳࢦఆ͕Ͱ͖Δ [ ] ͜ͷதʹ͋Δɾจࣈɾه߸ \\d \\D จࣈ . ͳΜͰO K * 0ճҎ্ͷ܁Γฦ͠ { } ׅހͷ͚ͩ܁Γฦ͠ ^ จࣈྻͷઌ಄ $ จࣈྻͷ࠷ޙ fi MFl.%$@@@@YMTY TUS@FYUSBDU fi MF z\^z TUS@FYUSBDU fi MF z.%$\^z TUS@FYUSBDU fi MF l<>Yz 4ܻ MDCɹҙͷ܁Γฦ͠ɹ4ܻ .(ه߸ͷυοτ)ɹ xɹɹɹҙͷ܁Γฦ͠
Slide 16
Slide 16 text
˔UJEZʢવσʔλʣʹ͢Δ ˔จࣈྻॲཧ ˔܁Γฦ͠ॲཧʹରԠ͢Δ ˔ؔͰ·ͱΊΔ 16
Slide 17
Slide 17 text
17 ྫʣ&YDFMͷෳλϒΛҰؾʹಡΈࠐΉ ܁Γฦ͠QVSSS
Slide 18
Slide 18 text
͚ͬͭ͘ΔɿEQMZSKPJO 18 https://github.com/rstudio/cheatsheets/raw/master/data-transformation.pdf
Slide 19
Slide 19 text
˔UJEZʢવσʔλʣʹ͢Δ ˔܁Γฦ͠ॲཧʹରԠ͢Δ ˔จࣈྻॲཧ ˔ؔͰ·ͱΊΔ 19
Slide 20
Slide 20 text
ؔͰ·ͱΊΔϝϦοτͱϙΠϯτ ˔ෆཁͳม͕ݮΔ ˔ଞͷϑΝΠϧͰར༻Ͱ͖ΔɻNBQ͕͍͘͢ͳΔɻ 20 ؔͰΘΕΔม Βͳ͍ ϑΥϧμʹ͋ΔϑΝΠϧΛ ·ͱΊͯಡΈࠐΊΔ
Slide 21
Slide 21 text
ؔͰ·ͱΊΔϝϦοτͱϙΠϯτ ˔͋Δ্͕ख͍͔ͬͨ͘Βͱ͍ͬͯҧ͏Ͱ্ख͘ߦ͔֬͘ೝඞཁ ɹɾYMT͕YMTYʹมΘͬͯରԠͰ͖Δίʔυ ɹɾΤϥʔ͕ग़ͨΒͦͷؔ͠ʹܧ͍͗ͯ͘͠ 21
Slide 22
Slide 22 text
۩ମྫ 22
Slide 23
Slide 23 text
ࠂࣔ൪߸Λଗ͑ΔͨΊͷલॲཧ 23
Slide 24
Slide 24 text
σʔλΛ֬ೝ ɾཉ͍͠σʔλ͚ͩ͜͜ ɾ˞ͷϚʔΫ͍Βͳ͍ ɾͷྻΛ͍ͨ͠ ɾվߦ͍Βͳ͍ 24 ▶︎ ͜Ε͚ͩͳΒίʔυ͔͚ͦ͏
Slide 25
Slide 25 text
Ͱྻ໊͕มΘ͍ͬͯͳ͍͔νΣοΫ 25 1ʙ3ྻͱ”ࢪઃ໊”ͷྻ͚ͩཉ͍͠ʂ select(1:3, “ࢪઃ໊”)
Slide 26
Slide 26 text
SFBE@FYDFMͰಡΈࠐΉ 26 read_excel ( path=ϑΝΠϧͷύε , sheet=γʔτͷ໊લ͔˓൪ͷγʔτ, #ࢦఆ͠ͳ͚Ε1ຕͷγʔτ skip=ઌ಄ʹ͍Βͳ͍ߦ͕͋Εࢦఆ , col_names=ྻ໊Λࢦఆ͍ͨ͠߹ʢFALSEʹ͢Δͱྻ໊ΛಡΈࠐ·ͳ͍ʣ , n_max=Կߦ·ͰಡΈࠐΉ͔ʁ ) ଓ͖σϞͰʂ
Slide 27
Slide 27 text
࡞ͬͨͷΛؔʹ͢Δ 27 ଓ͖σϞͰʂ
Slide 28
Slide 28 text
ࠂࣔ൪߸ͱ௨൪ͷؔ 28 ৽͍͠ͷ௨൪ͱݹ͍ͷࠂࣔ൪߸ͷࣈ͕߹͏Α͏ʹjoin͢Δ ৽͍͠ %>% left_join(ݹ͍, by = c(“௨൪” = “ࠂࣔ൪߸”)
Slide 29
Slide 29 text
ʂ 29 ଓ͖σϞͰʂ
Slide 30
Slide 30 text
ࢢொଜ൪߸ ૯লͷσʔλ͔ΒࢢொଜίʔυͷϑΝΠϧ͕μϯϩʔυͰ͖Δ 30 https://www.soumu.go.jp/denshijiti/code.html
Slide 31
Slide 31 text
ୀӃױऀௐࠪ 31 https://www.mhlw.go.jp/stf/shingi2/0000196043_00003.html
Slide 32
Slide 32 text
·ͣྻ໊͔Β 32 ෳྻ໊ ࡏӃ ͍Βͳ͍ ϋΠϑϯ ͍Βͳ͍ ͕มΘΔͱ ૿͑ͨΓݮͬͨΓ͢Δ ͕มΘΔͱ පӃ໊͕มΘΔ͜ͱ͕͋Δ ͕มΘΔͱ ප໊มΘΔ͜ͱ͕͋Δ ԣྻɺॎߦ ºؒ ʢ༌݂Ҏ֎ͷ࠶ܝʣ Λআ͍͕ͨ߹ܭ ͕มΘΔͱ ൪߸ҧ͏͜ͱ͕͋Δ ௨൪ʹલͷͷࠂࣔ൪߸
Slide 33
Slide 33 text
࡞ઓ ˔ྻ໊ͷΈΛಡΈࠐΈॎʹ͢Δ 33
Slide 34
Slide 34 text
࡞ઓ ˔UJEZS fi MM Ͱ/"Λ্ͷͰຒΊΔ 34
Slide 35
Slide 35 text
࡞ઓ ˔UJEZSVOJUF ͰෳྻΛͭʹ߹ମͤ͞Δ 35 ৽͍͠ྻ໊ ܨ͍͛ͨྻ ϕΫτϧʹม QVMM Ͱ0, ͜͜Ͱશ෦
Slide 36
Slide 36 text
࡞ઓ ˔TUS@SFQMBSF@BMMͰ/"ه߸Λফ͢ 36 ΈͰ͜ΕΒͷه߸εϖʔεΛফ͢͜ͱՄೳ ࠓճɻͱεϖʔεΛফͨ͠
Slide 37
Slide 37 text
ؔʹ͢Δ 37 ଓ͖σϞͰʂ
Slide 38
Slide 38 text
දه༳Εͷ֬ೝ 38 ࡏӃ ͍Βͳ͍ ϋΠϑϯ ͍Βͳ͍ ͕มΘΔͱ ૿͑ͨΓݮͬͨΓ͢Δ ͕มΘΔͱ පӃ໊͕มΘΔ͜ͱ͕͋Δ ͕มΘΔͱ ප໊มΘΔ͜ͱ͕͋Δ ԣྻɺॎߦ ºؒ ʢ༌݂Ҏ֎ͷ࠶ܝʣ Λআ͍͕ͨ߹ܭ ͕มΘΔͱ ൪߸ҧ͏͜ͱ͕͋Δ ௨൪ʹલͷͷࠂࣔ൪߸
Slide 39
Slide 39 text
දه༳ΕΛ୳͢ ˔֤ͷྻ໊Λऔಘ ˔TFUEJ ff Ͱยํʹ͔͠ͳ͍ͷΛ୳͢ TFUEJ ff " # ▶︎ "ʹ͔͠ແ͍ͷΛநग़ TFUEJ ff # " ▶︎ #ʹ͔͠ແ͍ͷΛநग़ 39
Slide 40
Slide 40 text
40 ͱɺͱಉ͡ ͱʹҧ͍͕͋Εදه༳ΕͷՄೳੑ දه༳ΕͰͳ͘ ͦͦͦͷʹ͔͠ͳ͔ͬͨՄೳੑ
Slide 41
Slide 41 text
͜ΕͰσʔλ͕ಡΈࠐΊΔʢޙͰ͢ʣ 41 ଓ͖σϞͰʂ
Slide 42
Slide 42 text
͍Βͳ͍ྻ 42 ࡏӃ ͍Βͳ͍ ϋΠϑϯ ͍Βͳ͍ ͕มΘΔͱ පӃ໊͕มΘΔ͜ͱ͕͋Δ ԣྻɺॎߦ ºؒ ʢ༌݂Ҏ֎ͷ࠶ܝʣ Λআ͍͕ͨ߹ܭ ͕มΘΔͱ ൪߸ҧ͏͜ͱ͕͋Δ ௨൪ʹલͷͷࠂࣔ൪߸ ͕มΘΔͱ ප໊มΘΔ͜ͱ͕͋Δ ͕มΘΔͱ ૿͑ͨΓݮͬͨΓ͢Δ
Slide 43
Slide 43 text
ࡏӃͱʢ࠶ܝʣͷྻ͍Βͳ͍ ˔TFMFDU ͱDPOUBJOT ͰߜΓࠐΊΔ 43
Slide 44
Slide 44 text
ϋΠϑϯΛফ͢ ▶︎ σʔλʹม 44 ϋΠϑϯ ͍Βͳ͍ ͕มΘΔͱ පӃ໊͕มΘΔ͜ͱ͕͋Δ ԣྻɺॎߦ ºؒ ͕มΘΔͱ ൪߸ҧ͏͜ͱ͕͋Δ ௨൪ʹલͷͷࠂࣔ൪߸ ͕มΘΔͱ ૿͑ͨΓݮͬͨΓ͢Δ ͕มΘΔͱ ප໊มΘΔ͜ͱ͕͋Δ
Slide 45
Slide 45 text
SFBESQBSTF@OVNCFS ˔SFBESQBSTF@OVNCFS ࢦఆͨ͠ྻΛσʔλʹ͢Δ ˔NVUBUF BDSPTT બͼ͍ͨྻ ؔ Ͱෳྻಉ࣌ʹॲཧ͢Δ 45
Slide 46
Slide 46 text
/"Λʹม͢Δ ˔SFQMBDF@OB ࢦఆͨ͠ྻ ม͍ͨ͠ ˔ؔʹҾ͕͋Δ߹ෳͷؔΛඞཁͱ͢Δ߹ ɹNVUBUF BDSPTT બͼ͍ͨྻ dؔ Λ͏ 46
Slide 47
Slide 47 text
͜͜·ͰΛؔʹ͢Δ 47 ଓ͖σϞͰʂ
Slide 48
Slide 48 text
ࠂࣔ൪߸͕Ͱҧ͏ 48 ͕มΘΔͱ පӃ໊͕มΘΔ͜ͱ͕͋Δ ԣྻɺॎߦ ºؒ ͕มΘΔͱ ൪߸ҧ͏͜ͱ͕͋Δ ௨൪ʹલͷͷࠂࣔ൪߸ ͕มΘΔͱ ૿͑ͨΓݮͬͨΓ͢Δ ͕มΘΔͱ ප໊มΘΔ͜ͱ͕͋Δ
Slide 49
Slide 49 text
ࢪઃ֓ཁදͱKPJO͢ΔʂDTWϑΝΠϧʹ͢Δ 49 ଓ͖σϞͰʂ
Slide 50
Slide 50 text
͜͜·ͰͰ͖Δͱ܁Γฦ͢͜ͱ͕Ͱ͖Δ 50 ԣྻɺॎߦ ºؒ ͕มΘΔͱ ૿͑ͨΓݮͬͨΓ͢Δ ͕มΘΔͱ ප໊มΘΔ͜ͱ͕͋Δ
Slide 51
Slide 51 text
51 ྫʣ&YDFMͷෳλϒΛҰؾʹಡΈࠐΉ ܁Γฦ͠QVSSS
Slide 52
Slide 52 text
˔ϑΥϧμʹ͋ΔΤΫηϧϑΝΠϧΛ·ͱΊͯಡΈࠐΉ -JTU fi MFT ϑΥϧμ໊ NBQؔΛͬͯ܁Γฦ͢ 52 pattern=Ͱ݅ΛߜΔ ଓ͖σϞͰʂ
Slide 53
Slide 53 text
ྻҧ͏ 53 ͕มΘΔͱ ૿͑ͨΓݮͬͨΓ͢Δ ͕มΘΔͱ ප໊มΘΔ͜ͱ͕͋Δ
Slide 54
Slide 54 text
XJEFσʔλ 54 σʔλ͕ԣʹฒΜͰ͍Δ
Slide 55
Slide 55 text
MPOHσʔλͳΒؔͳ͍ʂ ˔UJEZͳΒྻ͕૿͑Δ͚ͩ 55
Slide 56
Slide 56 text
UJEZSQJWPU@MPOHFSͰॎ࣋ͪσʔλʹม QJWPU@MPOHFS ɹDPMT·ͱΊ͍ͨྻ ɹOBNFT@UP·ͱ·ͬͨޙʢྻ໊ʣͷྻ໊ ɹWBMVFT@UP·ͱ·ͬͨޙʢʣͷྻ໊ 56
Slide 57
Slide 57 text
QJWPU@MPOHFSͷOBNFT@TFQ OBNFT@TFQͰྻΛׂ͢Δ͜ͱ͕Ͱ͖Δ 57
Slide 58
Slide 58 text
5JEZ 58 ଓ͖σϞͰʂ
Slide 59
Slide 59 text
දه༳ΕΛ͢ 59 ͕มΘΔͱ ප໊มΘΔ͜ͱ͕͋Δ
Slide 60
Slide 60 text
දه༳ΕΛ͢ ˔UJEZʹͳ͔ͬͨΒNVUBUFͱJGFMTFͰ͙͢ʹͤΔ 60
Slide 61
Slide 61 text
61 ྫʣ&YDFMͷෳλϒΛҰؾʹಡΈࠐΉ ܁Γฦ͠QVSSS
Slide 62
Slide 62 text
࠷ޙʹશ෦ͷϑΝΠϧΛಡΈࠐΉ 62 ଓ͖σϞͰʂ
Slide 63
Slide 63 text
͓ർΕ༷Ͱͨ͠ʂʂʂ ྻߦͷσʔλʂʂ 63
Slide 64
Slide 64 text
՝ ࠓճ࠷৽ͷͷපӃͷ໊લʹ߹Θ͍ͤͯΔ ɹɾ࠷৽ͷʹͳ͍පӃল͔Ε͍ͯΔ ɹɾ߹ซͨ͠පӃରԠ͍ͯ͠ͳ͍ ▶︎ ʢࣗͷࣄͷൣғͩͱࠔΒͳ͍͕ʣ ɹɹݚڀͰ͏ͳΒͬͱݫີʹ͢Δඞཁ͕͋Δ͔͠Εͳ͍ ▶︎ ɹ%1$Λѻ͏ํͷҙݟΛͬͯΈ͍ͨͰ͢ɻ 64
Slide 65
Slide 65 text
લॲཧ͕Ͱ͖Ε࣍ʹਐΊΔ ूܭɹɹɹɹɹɹɹɹɹɹɹɹɹɹ8ΞϓϦʢTIJOZʣ 65
Slide 66
Slide 66 text
TIJOZͳΒʂʂʂ 66
Slide 67
Slide 67 text
ऴΘΓʹ ˔&YDFMͷख࡞ۀͰͰ͖ͳ͍͜ͱ͕3QZUIPOͳΒͰ͖ͯ͠·͍·͢ ˔ҩྍ౷ܭ&YDFMͷ࡞Γํ͕ྑ͚Εલॲཧ΄΅ඞཁ͋Γ·ͤΜ ˔ͨͩ৭ʑͳσʔλΛੳͨ͘͠ͳΔͱલॲཧඞཁʹͳΓ·͢ ɹʢ༧ࢉͷׂҎ্લॲཧʹ͍ͬͯΔͱ͍͏ӟʜʣ ˔·ͩ·ͩஓͳίʔυଟ͍Ͱ͢ɻ ɹͲͷఔͷ࣮ྗ͔ʢશવμϝ ࠷ݶΫϦΞ ଈઓྗʹͳΔPSOPUʣ ɹΞυόΠεɾϑΟʔυόοΫ͍͚ͨͩΔͱ͍Ͱ͢ʂ 67