Slide 1

Slide 1 text

σʔλղੳͱલॲཧᶘ .ࠇ໦༟ୋ !FEUVTBDKQ

Slide 2

Slide 2 text

໨࣍  3FWJFX&YFSDJTF  +PJO  5JEZ%BUB !2

Slide 3

Slide 3 text

ຊ೔࢖༻͢Δσʔλ TUBSXBST w ελʔ΢Υʔζͷొ৔ਓ෺ʹؔ͢Δσʔλ IUUQTXBQJDP  qJHIUT w ೥ʹ-(" +', &83Λग़ൃͨ͢͠΂ͯͷϑϥΠτͷఆࠁσʔλ XFBUIFS w -(" +', &83ͷఱީ΍෩ͷ৘ใ ࣌ؒ͝ͱ  BJSMJOFT w ߤۭձࣾͷςʔϒϧ !3

Slide 4

Slide 4 text

3FWJFX&YFSDJTF

Slide 5

Slide 5 text

%BUB'SBNFͷجຊૢ࡞ EQMZS w ม਺ ྻ ͷநग़ w ؍ଌ ߦ ͷநग़ w ؍ଌ ߦ ͷฒͼସ͑ w ৽ͨͳม਺ ྻ ͷ࡞੒ w ूܭ w άϧʔϓԽ !5 • select() • filter() • arrange() • mutate() • summarise() • group_by()

Slide 6

Slide 6 text

࢖͍ํ w ୈҾ਺ʹ͸σʔλϑϨʔϜΛ༩͑Δ w ୈҾ਺Ҏ߱Ͱ͸ྻ໊ΛΫΦʔςʔγϣϯແ͠Ͱ༩͑Δ w ໭Γ஋͸৽ͨͳσʔλϑϨʔϜ %>%ͱ߹Θͤͯര଎σʔλϋϯυϦϯάʂʂ !6

Slide 7

Slide 7 text

ԋश qJHIUTσʔλʹؔͯ͠ɺҎԼͷ໰୊ʹ౴͑Α  ඈߦڑ཭͕࠷௕Ͱ͋Δศͷग़ൃ஍ͱ໨త஍͸Ͳ͔͜  ౸ண࣌ࠁͷ஗Ε͕ݦஶͳߤۭձࣾ͸Ͳ͔͜  ग़ൃ࣌ࠁͱ౸ண࣌ࠁͷ஗Ε͕ݦஶͳߤۭձࣾ͸Ͳ͔͜  Կ࣌ൃͷඈߦػ͕࠷΋ଟ͍͔  ߤۭձࣾͷൟ๩ظ͸͍͔ͭ  શͯͷߦͰdep_time - sched_dep_time = dep_delayͱͳ͍ͬͯΔ͜ͱΛ֬ೝ ͤΑ !7 # ύοέʔδ͔ΒಡΈࠐΉ library(nycflights13) data(flights)

Slide 8

Slide 8 text

+PJO

Slide 9

Slide 9 text

+PJO ͭͷςʔϒϧΛ LFZΛ΋ͱʹ݁߹͢Δૢ࡞ w ʮֶੜͷݸਓ৘ใςʔϒϧʯ w ʮतۀͷ৘ใςʔϒϧʯ w ʮཤमɾ੒੷ςʔϒϧʯ LFZ w ʮֶੜʯ ʮ੒੷ʯɿLFZ͸ֶ੶൪߸ w ʮतۀʯ ʮཤमʯɿLFZ͸तۀ*% !9 ʮਓɾतۀɾ੒੷ͷςʔϒϧʯ

Slide 10

Slide 10 text

+PJOͷछྨ w YͱZΛ+PJO͍ͨ͠ w ΋ͬͱ΋୯७ͳͷ͸ *OOFSKPJO w ॏෳ͢ΔLFZ͚ͩ࢒͢ !10 ग़యɿIUUQTSETIBEDPO[

Slide 11

Slide 11 text

w -FGUKPJO w YͷLFZΛશͯ࢒͢ w 3JHIUKPJO w ZͷLFZΛશͯ࢒͢ w 'VMMKPJO w ྆ํͷLFZΛશͯ࢒͢ !11 ग़యɿIUUQTSETIBEDPO[

Slide 12

Slide 12 text

**_join()ͷ࢖͍ํ inner_join(band_members, band_instruments,
 by = “name”) left_join(band_members, band_instruments2,
 by = c(“name” = “artist”)) !12 > band_members name band 1 Mick Stones 2 John Beatles 3 Paul Beatles > band_instruments name plays 1 John guitar 2 Paul bass 3 Keith guitar > band_instruments2 artist plays 1 John guitar 2 Paul bass 3 Keith guitar

Slide 13

Slide 13 text

࿅श໰୊  inner_join(), left_join(), right_join(), full_join()
 ͦΕͧΕͷग़ྗ݁ՌΛ༧૝͠ ࣮ࡍʹಈ͔ͯ֬͠ೝͤΑ  qJHIUTσʔλͱBJSMJOFTσʔλΛDBSSJFSྻͰ݁߹ͤΑ  qJHIUTσʔλͱXFBUIFSσʔλΛPSJHJO ZFBS NPOUI EBZ IPVS ྻͰ݁߹ͤΑ !13

Slide 14

Slide 14 text

5JEZ%BUB

Slide 15

Slide 15 text

UJEZEBUB ͖ͪΜͱͨ͠σʔλ ఆٛʢग़యɿIUUQTSETIBEDPO[ʣ w Ұͭͷྻʹ͸Ұͭͷม਺ BUPNJDWFDUPS  w Ұͭͷߦʹ͸Ұͭͷ؍ଌ w Ұͭͷηϧʹ͸Ұͭͷ஋ w ݸʑͷ؍ଌ͸શͯಉ͡ܗΛ͍ͯ͠Δ σʔλϑϨʔϜ͸্هΛຬͨ͢Α͏ʹ࡞Ζ͏ ˞ߦ໊ʢSPXOBNFTʣ͸࢖ΘͣʹJOEFY΍JEͷྻΛ࡞Ζ͏ !15

Slide 16

Slide 16 text

NFTTZEBUB w Α͘ݟΔܗ w ਓؒʹ͸Θ͔Γ΍͍͢ ʮԣ࣋ͪܗʯ w Ұͭͷྻʹ͸Ұͭͷม਺˚ w Ұͭͷߦʹ͸Ұͭͷ؍ଌ✖ w Ұͭͷηϧʹ͸Ұͭͷ஋̋ !16 ஍఺ 12࣌ 15࣌ 17࣌ ౦ژ ‗ ‘ ‘ ໊ݹ԰ ‗ ‗ ‘ େࡕ ‘ ‘ ‘ ྻ໊ ߦ໊

Slide 17

Slide 17 text

NFTTZEBUB w Α͘ݟΔܗ w ਓؒʹ͸Θ͔Γ΍͍͢ ʮԣ࣋ͪܗʯ w Ұͭͷྻʹ͸Ұͭͷม਺˚ w Ұͭͷߦʹ͸Ұͭͷ؍ଌ✖ w Ұͭͷηϧʹ͸Ұͭͷ஋̋ !17 ஍఺ 12࣌ 15࣌ 17࣌ ౦ژ ‗ ‘ ‘ ໊ݹ԰ ‗ ‗ ‘ େࡕ ‘ ‘ ‘ ஍఺ ࣌ࠁ ఱؾ

Slide 18

Slide 18 text

UJEZEBUB w ղੳͰѻ͍΍͍͢ w ׳Εͳ͍͏ͪ͸ݟʹ͍͘ʁ ʮॎ࣋ͪܗʯ w Ұͭͷྻʹ͸Ұͭͷม਺̋ w Ұͭͷߦʹ͸Ұͭͷ؍ଌ̋ w Ұͭͷηϧʹ͸Ұͭͷ஋̋ !18 ஍఺ ࣌ࠁ ఱؾ ౦ژ ࣌ ‗ ໊ݹ԰ ࣌ ‗ େࡕ ࣌ ‘ ౦ژ ࣌ ‘ ໊ݹ԰ ࣌ ‗ େࡕ ࣌ ‘

Slide 19

Slide 19 text

NFTTZŠUJEZ !19  ྻ໊ʹͳͬͯ͠·͍ͬͯͨม਺໊   Λ
 ৽͍͠ZFBSͱ͍͏ม਺ʹ͢Δ

Slide 20

Slide 20 text

UJEZŠNFTTZ !20 

Slide 21

Slide 21 text

3Ͱͷॎԣม׵ !21 ॎ࣋ͪ ԣ࣋ͪ spread() gather() gather(df, key = “ྻ໊ʹདྷ͍ͯͨม਺Λ֨ೲ͢Δ৽ͨͳม਺໊”, value = “ෳ਺ͷྻʹ·͕͍ͨͬͯͨม਺Λ·ͱΊΔ৽ͨͳม਺໊”, - ม׵ʹߟྀ͠ͳ͍ྻ໊) spread(df, key, value, fill = ޿͛ͨͱ͖ܽଌʹͳΔͱ͜ΖΛຒΊ͍ͨ஋)

Slide 22

Slide 22 text

࿅श໰୊  ҎԼͷίʔυͰTUPDLT ٖࣅతͳऩӹ཰σʔλ Λ࡞Γ  ॎ௕ʹͤΑ stocks <- data.frame( time = as.Date('2009-01-01') + 0:9, X = rnorm(10, 0, 1), Y = rnorm(10, 0, 2), Z = rnorm(10, 0, 4) )  ΋ͱʹ໭ͤ !22

Slide 23

Slide 23 text

࣍ճ·Ͱͷ՝୊

Slide 24

Slide 24 text

՝୊ 1. ࠷΋ؾԹ͕ߴ͍தग़ൃͨ͠ศΛ೺ѲͤΑ 2. ଌఆ͞Εͨσʔλͷ͏ͪɺϘʔΠϯάࣾͷඈߦػ͸ԿճඈΜͰ͍Δ͔ 3. ඈߦػʹ࠾༻͞Ε͍ͯΔΤϯδϯͷछྨ͝ͱʹɺ1ճ͋ͨΓͷฏۉඈ ߦڑ཭Λࢉग़ͤΑ 4. ୹ڑ཭ or ௕ڑ཭ʹಛԽ͍ͯ͠Δߤۭձࣾ͸͋Δ͔ɻ͋ΔͳΒ͹ɺ൑அ ཧ༝΋ड़΂Αɻ 5. ౦ʹ޲͔ͬͯඈͿศͱ੢ʹ޲͔ͬͯඈͿศͷͲͪΒ͕ଟ͍͔ (ඈߦػ ͸໨త஍ʹ޲͔ͬͯ௚ਐ͢Δ΋ͷͱ͢Δ) 6. ग़ൃ࣌ͷ࣪౓ͱɺग़ൃͷ஗Ԇʹ૬ؔ͸͋Δ͔ !24

Slide 25

Slide 25 text

Α͋͘Δ࣭໰ w σʔλαΠΤϯεͷԿָ͕͍͠ʁ w σʔλ͔Β஌ݟΛಘΔ ͱ͍͏खଓ͖͕ԿΑΓ΋ָ͍͠ ࢲݟ  w Ծઆɾݕূ͕ΩϨΠʹܾ·ͬͨͱ͖͕ؾ͍͍࣋ͪ w ೥ੜͷ͏ͪ͸ԿΛͨ͠Βྑ͍ʁ w جૅ ౷ܭֶ ࠷దԽ ઢܗ୅਺ FUD ΛΩϟονΞοϓ͢Δ࣌ؒ͸ࠓޙͳ͘ͳͬͯ ͍͘ w ڵຯͷ͋Δσʔλ ڝഅ εϙʔπ FUD Λର৅ʹ ෼ੳΛֶΜͰ͍͘ͷ΋ྑ͍͔ ΋ָ͠Ήͷ͕Ұ൪ w 3͕೉͍͠ w ؆୯΍ͦ͞͠͏ͳࢀߟॻΛݟͯΈΔͷ΋˕ !25