Как свернуть строки в аккуратном мире или с помощью data.table после распространения данных long -> wide - PullRequest
Новая
       52

Как свернуть строки в аккуратном мире или с помощью data.table после распространения данных long -> wide

1 голос
/ 12 апреля 2019

Я работаю в колл-центре, и эти данные - все соответствующие данные для нашей системы автодозвона. Начиная с 2019 года наша компания запустила несколько кампаний, чтобы охватить участников множеством возможных рецептов вместо 1.

В настоящее время мы сталкиваемся с тем, что у члена X будет потенциально 6 рецептов, с которыми мы пытаемся связаться с ними, и зачастую они дублируются. Отдел, собирающий все эти данные, просматривает столбцы IDCARD_MBR_ID и TARGET_DOC_ для уникальных значений, не обращая внимания на название рецептурного препарата.

Моя цель - удалить дубликаты отпускаемых по рецепту лекарств. и пополнить в любом свободном месте

Порядок приема лекарств также не всегда одинаков. Краткий пример этого может выглядеть следующим образом: 

H0000000001 (member data columns) Target1, Target2 
H0000000001 (same member data)    Target3, Target1, Target2

я бы хотел видеть: 

H0000000001 (member data columns) Target1, Target2,Target3

Я уже обработал свои данные с помощью сбора, удалил дубликаты, перенумеровал свои цели 1-6 (чтобы я мог создать новую строку, если найдено> 3) и распространил ее обратно. Моя проблема с моим распространением .. У меня есть нечто, похожее на это: 

structure(list(IDCARD_MBR_ID = c("H0000000001", "H0000000001", 
"H0000000001", "H0000000002", "H0000000002", "H0000000002", "H0000000003", 
"H0000000003", "H0000000003", "H0000000004", "H0000000004", "H0000000004", 
"H0000000005", "H0000000005", "H0000000005", "H0000000006", "H0000000006", 
"H0000000006", "H0000000007", "H0000000007", "H0000000007", "H0000000008", 
"H0000000008", "H0000000008", "H0000000009", "H0000000009", "H0000000009", 
"H0000000010", "H0000000010", "H0000000010", "H0000000011", "H0000000011", 
"H0000000011", "H0000000012", "H0000000012", "H0000000012"), 
    SDR_PERSON_ID = c("ID000000001", "ID000000001", "ID000000001", 
    "ID000000002", "ID000000002", "ID000000002", "ID000000003", 
    "ID000000003", "ID000000003", "ID000000004", "ID000000004", 
    "ID000000004", "ID000000005", "ID000000005", "ID000000005", 
    "ID000000006", "ID000000006", "ID000000006", "ID000000007", 
    "ID000000007", "ID000000007", "ID000000008", "ID000000008", 
    "ID000000008", "ID000000009", "ID000000009", "ID000000009", 
    "ID000000010", "ID000000010", "ID000000010", "ID000000011", 
    "ID000000011", "ID000000011", "ID000000012", "ID000000012", 
    "ID000000012"), FIRST_NAME = c("First1", "First1", "First1", 
    "First2", "First2", "First2", "First3", "First3", "First3", 
    "First4", "First4", "First4", "First5", "First5", "First5", 
    "First6", "First6", "First6", "First7", "First7", "First7", 
    "First8", "First8", "First8", "First9", "First9", "First9", 
    "First10", "First10", "First10", "First11", "First11", "First11", 
    "First12", "First12", "First12"), LAST_NAME = c("Last1", 
    "Last1", "Last1", "Last2", "Last2", "Last2", "Last3", "Last3", 
    "Last3", "Last4", "Last4", "Last4", "Last5", "Last5", "Last5", 
    "Last6", "Last6", "Last6", "Last7", "Last7", "Last7", "Last8", 
    "Last8", "Last8", "Last9", "Last9", "Last9", "Last10", "Last10", 
    "Last10", "Last11", "Last11", "Last11", "Last12", "Last12", 
    "Last12"), GENDER = c("F", "F", "F", "M", "M", "M", "M", 
    "M", "M", "F", "F", "F", "F", "F", "F", "F", "F", "F", "M", 
    "M", "M", "F", "F", "F", "F", "F", "F", "M", "M", "M", "F", 
    "F", "F", "F", "F", "F"), DOB = c("1901-01-01", "1901-01-01", 
    "1901-01-01", "1901-01-02", "1901-01-02", "1901-01-02", "1901-01-03", 
    "1901-01-03", "1901-01-03", "1901-01-04", "1901-01-04", "1901-01-04", 
    "1901-01-05", "1901-01-05", "1901-01-05", "1901-01-06", "1901-01-06", 
    "1901-01-06", "1901-01-07", "1901-01-07", "1901-01-07", "1901-01-08", 
    "1901-01-08", "1901-01-08", "1901-01-09", "1901-01-09", "1901-01-09", 
    "1901-01-10", "1901-01-10", "1901-01-10", "1901-01-11", "1901-01-11", 
    "1901-01-11", "1901-01-12", "1901-01-12", "1901-01-12"), 
    ADDRESS1 = c("123 Testing Way", "123 Testing Way", "123 Testing Way", 
    "456 Testing Way", "456 Testing Way", "456 Testing Way", 
    "789 Testing Way", "789 Testing Way", "789 Testing Way", 
    "1122 Testing Way", "1122 Testing Way", "1122 Testing Way", 
    "1455 Testing Way", "1455 Testing Way", "1455 Testing Way", 
    "1788 Testing Way", "1788 Testing Way", "1788 Testing Way", 
    "2121 Testing Way", "2121 Testing Way", "2121 Testing Way", 
    "2454 Testing Way", "2454 Testing Way", "2454 Testing Way", 
    "2787 Testing Way", "2787 Testing Way", "2787 Testing Way", 
    "3120 Testing Way", "3120 Testing Way", "3120 Testing Way", 
    "3453 Testing Way", "3453 Testing Way", "3453 Testing Way", 
    "3786 Testing Way", "3786 Testing Way", "3786 Testing Way"
    ), ADDRESS2 = c(NA, NA, NA, NA, NA, NA, "APT 302", "APT 302", 
    "APT 302", "APT 225", "APT 225", "APT 225", NA, NA, NA, NA, 
    NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
    NA, NA, NA, NA, NA), CITY = c("CLARKSVILLE", "CLARKSVILLE", 
    "CLARKSVILLE", "MONROE", "MONROE", "MONROE", "LOS ANGELES", 
    "LOS ANGELES", "LOS ANGELES", "TULSA", "TULSA", "TULSA", 
    "RED BUD", "RED BUD", "RED BUD", "MISSION VIEJO", "MISSION VIEJO", 
    "MISSION VIEJO", "MILLCREEK", "MILLCREEK", "MILLCREEK", "COLUMBIA", 
    "COLUMBIA", "COLUMBIA", "HOUSTON", "HOUSTON", "HOUSTON", 
    "APOPKA", "APOPKA", "APOPKA", "SEALY", "SEALY", "SEALY", 
    "SEWARD", "SEWARD", "SEWARD"), ST = c("IN", "IN", "IN", "MI", 
    "MI", "MI", "CA", "CA", "CA", "OK", "OK", "OK", "IL", "IL", 
    "IL", "CA", "CA", "CA", "UT", "UT", "UT", "MO", "MO", "MO", 
    "TX", "TX", "TX", "FL", "FL", "FL", "TX", "TX", "TX", "NE", 
    "NE", "NE"), ZIP = c(47129L, 47129L, 47129L, 48162L, 48162L, 
    48162L, 90028L, 90028L, 90028L, 74136L, 74136L, 74136L, 62278L, 
    62278L, 62278L, 92692L, 92692L, 92692L, 84124L, 84124L, 84124L, 
    65201L, 65201L, 65201L, 77057L, 77057L, 77057L, 32703L, 32703L, 
    32703L, 77474L, 77474L, 77474L, 68434L, 68434L, 68434L), 
    EMAIL_ADDRESS = c("Test-mail@test.com", "Test-mail@test.com", 
    "Test-mail@test.com", NA, NA, NA, NA, NA, NA, NA, NA, NA, 
    "Test-mail@test.com", "Test-mail@test.com", "Test-mail@test.com", 
    NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, "Test-mail@test.com", 
    "Test-mail@test.com", "Test-mail@test.com", NA, NA, NA, NA, 
    NA, NA), PHONE1 = c(1L, 1L, 1L, 2L, 2L, 2L, 3L, 3L, 3L, 4L, 
    4L, 4L, 5L, 5L, 5L, 6L, 6L, 6L, 7L, 7L, 7L, 8L, 8L, 8L, 9L, 
    9L, 9L, 10L, 10L, 10L, 11L, 11L, 11L, 12L, 12L, 12L), ARG_CLIENT_NUM = c(6L, 
    6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 
    6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 
    6L, 6L, 6L, 6L, 6L), FORMULARYID = c(19430L, 19430L, 19431L, 
    19431L, 19431L, 19430L, 19430L, 19430L, 19430L, 19430L, 19430L, 
    19430L, 19430L, 19430L, 19430L, 19431L, 19431L, 19431L, 19430L, 
    19430L, 19431L, 19431L, 19431L, 19430L, 19430L, 19430L, 19431L, 
    19430L, 19430L, 19430L, 19430L, 19430L, 19430L, 19430L, 19430L, 
    19431L), CAMPAIGN_TYPE = c("PDP_Lean MAPD", "PDP_Lean MAPD", 
    "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", 
    "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", 
    "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", 
    "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", 
    "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", 
    "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", 
    "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", 
    "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", 
    "PDP_Lean MAPD", "PDP_Lean MAPD"), LIS_INDICATOR = c("N", 
    "N", "N", "N", "N", "Y", "Y", "Y", "Y", "Y", "Y", "Y", "Y", 
    "Y", "N", "N", "N", "N", "Y", "Y", "N", "N", "N", "N", "N", 
    "N", "N", "Y", "Y", "N", "Y", "Y", "Y", "Y", "Y", "N"), MONTH = c("February / DIALER", 
    "February / DIALER", "January / DIALER", "January / DIALER", 
    "January / DIALER", "January / DIALER", "February / DIALER", 
    "February / DIALER", "January / DIALER", "January / DIALER", 
    "January / DIALER", "January / DIALER", "January / DIALER", 
    "January / DIALER", "January / DIALER", "January / DIALER", 
    "January / DIALER", "January / DIALER", "January / DIALER", 
    "January / DIALER", "February / DIALER", "January / DIALER", 
    "January / DIALER", "February / DIALER", "January / DIALER", 
    "January / DIALER", "January / DIALER", "January / DIALER", 
    "January / DIALER", "February / DIALER", "January / DIALER", 
    "January / DIALER", "January / DIALER", "January / DIALER", 
    "January / DIALER", "January / DIALER"), Calltype1 = c("Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings"), Calltype2 = c("Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings"), Calltype3 = c("Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings"), TARGET_DOC_1 = c("A000000000025001", 
    NA, NA, "A000000000026001", NA, NA, "A000000000027001", NA, 
    NA, "A000000000028001", NA, NA, "A000000000029001", NA, NA, 
    "A000000000030001", NA, NA, "A000000000031001", NA, NA, "A000000000032001", 
    NA, NA, "A000000000033001", NA, NA, "A000000000034001", NA, 
    NA, "A000000000035001", NA, NA, "A000000000036001", NA, NA
    ), TARGET_DOC_2 = c(NA, "A000000000013001", NA, NA, "A000000000014001", 
    NA, NA, "A000000000015001", NA, NA, "A000000000016001", NA, 
    NA, "A000000000017001", NA, NA, "A000000000018001", NA, NA, 
    "A000000000019001", NA, NA, "A000000000020001", NA, NA, "A000000000021001", 
    NA, NA, "A000000000022001", NA, NA, "A000000000023001", NA, 
    NA, "A000000000024001", NA), TARGET_DOC_3 = c(NA, NA, "A000000000001001", 
    NA, NA, "A000000000002001", NA, NA, "A000000000003001", NA, 
    NA, "A000000000004001", NA, NA, "A000000000005001", NA, NA, 
    "A000000000006001", NA, NA, "A000000000007001", NA, NA, "A000000000008001", 
    NA, NA, "A000000000009001", NA, NA, "A000000000010001", NA, 
    NA, "A000000000011001", NA, NA, "A000000000012001"), TARGET_DRUG_ID_1 = c(57237014401, 
    NA, NA, 65862058301, NA, NA, 57237014301, NA, NA, 68180075701, 
    NA, NA, 65862058401, NA, NA, 55111033901, NA, NA, 65862058201, 
    NA, NA, 65862058405, NA, NA, 65862058601, NA, NA, 68180075801, 
    NA, NA, 65862058601, NA, NA, 68180076001, NA, NA), TARGET_DRUG_ID_2 = c(NA, 
    57237014401, NA, NA, 65862058301, NA, NA, 57237014301, NA, 
    NA, 68180075701, NA, NA, 65862058401, NA, NA, 55111033901, 
    NA, NA, 65862058201, NA, NA, 65862058405, NA, NA, 65862058601, 
    NA, NA, 68180075801, NA, NA, 65862058601, NA, NA, 68180076001, 
    NA), TARGET_DRUG_ID_3 = c(NA, NA, 65862058701, NA, NA, 65862058701, 
    NA, NA, 55111034001, NA, NA, 65862058601, NA, NA, 65862058701, 
    NA, NA, 55111034001, NA, NA, 65862058401, NA, NA, 65862058405, 
    NA, NA, 68180047301, NA, NA, 68180075801, NA, NA, 65862058701, 
    NA, NA, 65862058301), TARGET_NAME_1 = c("Target11", NA, NA, 
    "Target12", NA, NA, "Target13", NA, NA, "Target14", NA, NA, 
    "Target15", NA, NA, "Target16", NA, NA, "Target17", NA, NA, 
    "Target18", NA, NA, "Target19", NA, NA, "Target20", NA, NA, 
    "Target21", NA, NA, "Target22", NA, NA), TARGET_NAME_2 = c(NA, 
    "Target21", NA, NA, "Target22", NA, NA, "Target23", NA, NA, 
    "Target24", NA, NA, "Target25", NA, NA, "Target26", NA, NA, 
    "Target27", NA, NA, "Target28", NA, NA, "Target29", NA, NA, 
    "Target30", NA, NA, "Target31", NA, NA, "Target32", NA), 
    TARGET_NAME_3 = c(NA, NA, "Target31", NA, NA, "Target32", 
    NA, NA, "Target33", NA, NA, "Target34", NA, NA, "Target35", 
    NA, NA, "Target36", NA, NA, "Target37", NA, NA, "Target38", 
    NA, NA, "Target39", NA, NA, "Target40", NA, NA, "Target41", 
    NA, NA, "Target42")), class = "data.frame", row.names = c(NA, 
-36L))}

Каждый набор целей находится на отдельной линии ... и я не могу понять, как их свернуть и удалить NA, чтобы я получил результат, который выглядит следующим образом: 

structure(list(IDCARD_MBR_ID = c("H0000000001", "H0000000002", 
"H0000000003", "H0000000004", "H0000000005", "H0000000006", "H0000000007", 
"H0000000008", "H0000000009", "H0000000010", "H0000000011", "H0000000012"
), SDR_PERSON_ID = c("ID000000001", "ID000000002", "ID000000003", 
"ID000000004", "ID000000005", "ID000000006", "ID000000007", "ID000000008", 
"ID000000009", "ID000000010", "ID000000011", "ID000000012"), 
    FIRST_NAME = c("First1", "First2", "First3", "First4", "First5", 
    "First6", "First7", "First8", "First9", "First10", "First11", 
    "First12"), LAST_NAME = c("Last1", "Last2", "Last3", "Last4", 
    "Last5", "Last6", "Last7", "Last8", "Last9", "Last10", "Last11", 
    "Last12"), GENDER = c("F", "M", "M", "F", "F", "F", "M", 
    "F", "F", "M", "F", "F"), DOB = c("1/1/1901", "1/2/1901", 
    "1/3/1901", "1/4/1901", "1/5/1901", "1/6/1901", "1/7/1901", 
    "1/8/1901", "1/9/1901", "1/10/1901", "1/11/1901", "1/12/1901"
    ), ADDRESS1 = c("123 Testing Way", "456 Testing Way", "789 Testing Way", 
    "1122 Testing Way", "1455 Testing Way", "1788 Testing Way", 
    "2121 Testing Way", "2454 Testing Way", "2787 Testing Way", 
    "3120 Testing Way", "3453 Testing Way", "3786 Testing Way"
    ), ADDRESS2 = c(NA, NA, "APT 302", "APT 225", NA, NA, NA, 
    NA, NA, NA, NA, NA), CITY = c("CLARKSVILLE", "MONROE", "LOS ANGELES", 
    "TULSA", "RED BUD", "MISSION VIEJO", "MILLCREEK", "COLUMBIA", 
    "HOUSTON", "APOPKA", "SEALY", "SEWARD"), ST = c("IN", "MI", 
    "CA", "OK", "IL", "CA", "UT", "MO", "TX", "FL", "TX", "NE"
    ), ZIP = c(47129L, 48162L, 90028L, 74136L, 62278L, 92692L, 
    84124L, 65201L, 77057L, 32703L, 77474L, 68434L), EMAIL_ADDRESS = c("Test-mail@test.com", 
    NA, NA, NA, "Test-mail@test.com", NA, NA, NA, NA, "Test-mail@test.com", 
    NA, NA), PHONE1 = 1:12, ARG_CLIENT_NUM = c(6L, 6L, 6L, 6L, 
    6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L), FORMULARYID = c(19430L, 
    19431L, 19430L, 19430L, 19430L, 19431L, 19430L, 19431L, 19430L, 
    19430L, 19430L, 19430L), CAMPAIGN_TYPE = c("PDP_Lean MAPD", 
    "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", 
    "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD", 
    "PDP_Lean MAPD", "PDP_Lean MAPD", "PDP_Lean MAPD"), LIS_INDICATOR = c("N", 
    "N", "Y", "Y", "Y", "N", "Y", "N", "N", "Y", "Y", "Y"), MONTH = c("February / DIALER", 
    "January / DIALER", "February / DIALER", "January / DIALER", 
    "January / DIALER", "January / DIALER", "January / DIALER", 
    "January / DIALER", "January / DIALER", "January / DIALER", 
    "January / DIALER", "January / DIALER"), Calltype1 = c("Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings"), Calltype2 = c("Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings"), Calltype3 = c("Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings", "Cost Savings", 
    "Cost Savings", "Cost Savings", "Cost Savings"), TARGET_DOC_1 = c("A000000000025001", 
    "A000000000026001", "A000000000027001", "A000000000028001", 
    "A000000000029001", "A000000000030001", "A000000000031001", 
    "A000000000032001", "A000000000033001", "A000000000034001", 
    "A000000000035001", "A000000000036001"), TARGET_DOC_2 = c("A000000000013001", 
    "A000000000014001", "A000000000015001", "A000000000016001", 
    "A000000000017001", "A000000000018001", "A000000000019001", 
    "A000000000020001", "A000000000021001", "A000000000022001", 
    "A000000000023001", "A000000000024001"), TARGET_DOC_3 = c("A000000000001001", 
    "A000000000002001", "A000000000003001", "A000000000004001", 
    "A000000000005001", "A000000000006001", "A000000000007001", 
    "A000000000008001", "A000000000009001", "A000000000010001", 
    "A000000000011001", "A000000000012001"), TARGET_DRUG_ID_1 = c(57237014401, 
    65862058301, 57237014301, 68180075701, 65862058401, 55111033901, 
    65862058201, 65862058405, 65862058601, 68180075801, 65862058601, 
    68180076001), TARGET_DRUG_ID_2 = c(57237014401, 65862058301, 
    57237014301, 68180075701, 65862058401, 55111033901, 65862058201, 
    65862058405, 65862058601, 68180075801, 65862058601, 68180076001
    ), TARGET_DRUG_ID_3 = c(65862058701, 65862058701, 55111034001, 
    65862058601, 65862058701, 55111034001, 65862058401, 65862058405, 
    68180047301, 68180075801, 65862058701, 65862058301), TARGET_NAME_1 = c("Target11", 
    "Target12", "Target13", "Target14", "Target15", "Target16", 
    "Target17", "Target18", "Target19", "Target20", "Target21", 
    "Target22"), TARGET_NAME_2 = c("Target21", "Target22", "Target23", 
    "Target24", "Target25", "Target26", "Target27", "Target28", 
    "Target29", "Target30", "Target31", "Target32"), TARGET_NAME_3 = c("Target31", 
    "Target32", "Target33", "Target34", "Target35", "Target36", 
    "Target37", "Target38", "Target39", "Target40", "Target41", 
    "Target42")), class = "data.frame", row.names = c(NA, -12L
))

Я смотрел на подобные вопросы, такие как этот: Свернуть дублированные строки Свернуть дублированные строки с разными значениями в нескольких столбцах, используя R

Моя самая большая проблема в том, что число моих столбцов намного больше (по необходимости ... для целостности все строки должны оставаться вместе). Я очень быстро теряюсь.

РЕДАКТИРОВАТЬ 1: Добавление 'head (12)' DF (обратите внимание, что DF очень широкий и его трудно вставить для форматирования. Вот столбцы, затронутые проблемой) ... также, не уверен, почему в этом тестовые данные, некоторые из которых являются NA, а некоторые являются NA .. в моих реальных данных они являются истинными пропущенными значениями NA. 


     Calltype3     TARGET_DOC_1     TARGET_DOC_2     TARGET_DOC_3 TARGET_DRUG_ID_1 TARGET_DRUG_ID_2 TARGET_DRUG_ID_3
1  Cost Savings A000000000025001             <NA>             <NA>      57237014401               NA               NA
2  Cost Savings             <NA> A000000000013001             <NA>               NA      57237014401               NA
3  Cost Savings             <NA>             <NA> A000000000001001               NA               NA      65862058701
4  Cost Savings A000000000026001             <NA>             <NA>      65862058301               NA               NA
5  Cost Savings             <NA> A000000000014001             <NA>               NA      65862058301               NA
6  Cost Savings             <NA>             <NA> A000000000002001               NA               NA      65862058701
7  Cost Savings A000000000027001             <NA>             <NA>      57237014301               NA               NA
8  Cost Savings             <NA> A000000000015001             <NA>               NA      57237014301               NA
9  Cost Savings             <NA>             <NA> A000000000003001               NA               NA      55111034001
10 Cost Savings A000000000028001             <NA>             <NA>      68180075701               NA               NA
11 Cost Savings             <NA> A000000000016001             <NA>               NA      68180075701               NA
12 Cost Savings             <NA>             <NA> A000000000004001               NA               NA      65862058601
   TARGET_NAME_1 TARGET_NAME_2 TARGET_NAME_3
1       Target11          <NA>          <NA>
2           <NA>      Target21          <NA>
3           <NA>          <NA>      Target31
4       Target12          <NA>          <NA>
5           <NA>      Target22          <NA>
6           <NA>          <NA>      Target32
7       Target13          <NA>          <NA>
8           <NA>      Target23          <NA>
9           <NA>          <NA>      Target33
10      Target14          <NA>          <NA>
11          <NA>      Target24          <NA>
12          <NA>          <NA>      Target34

РЕДАКТИРОВАТЬ 4: на основе предложения Humpelstielzchen я выполнил это: Мой спред DF = spread1 

spread2 <- spread1 %>% 
  group_by(IDCARD_MBR_ID) %>% 
  mutate(MONTH = MONTH[row_number()==1],
         FORMULARYID = FORMULARYID[row_number()==1],
         LIS_INDICATOR = LIS_INDICATOR[row_number() ==1]) %>% 
  group_by_at(vars(-starts_with("TARGET"))) %>% 
  summarise_at(vars(starts_with("TARGET")), list(. = ~.[which(!is.na(.))[1]]))

и, похоже, он работал правильно; У меня есть фрейм данных с более чем 1 миллионом начальных строк, это помогло мне сократить до ~ 600k; мне потребуется немного времени, чтобы проверить точность, но выглядит великолепно!

1 Ответ

3 голосов
/ 12 апреля 2019

Попробуйте это: 

library(dplyr)


df1 %>%
  group_by(IDCARD_MBR_ID) %>%
  mutate(MONTH = MONTH[row_number() == 1],
         FORMULARYID = FORMULARYID[row_number() == 1],
         LIS_INDICATOR = LIS_INDICATOR[row_number() == 1]) %>%
  group_by_at(vars(-starts_with("TARGET"))) %>%
  summarise_at(vars(starts_with("TARGET")), list(. = ~.[which(!is.na(.))[1]])) -> res


> res[,c("IDCARD_MBR_ID", "TARGET_NAME_1_.", "TARGET_NAME_2_.", "TARGET_NAME_3_." )]
# A tibble: 12 x 4
   IDCARD_MBR_ID TARGET_NAME_1_. TARGET_NAME_2_. TARGET_NAME_3_.
   <chr>         <chr>           <chr>           <chr>          
 1 H0000000001   Target11        Target21        Target31       
 2 H0000000002   Target12        Target22        Target32       
 3 H0000000003   Target13        Target23        Target33       
 4 H0000000004   Target14        Target24        Target34       
 5 H0000000005   Target15        Target25        Target35       
 6 H0000000006   Target16        Target26        Target36       
 7 H0000000007   Target17        Target27        Target37       
 8 H0000000008   Target18        Target28        Target38       
 9 H0000000009   Target19        Target29        Target39       
10 H0000000010   Target20        Target30        Target40       
11 H0000000011   Target21        Target31        Target41       
12 H0000000012   Target22        Target32        Target42

Мне пришлось выровнять некоторые группирующие переменные (MONTH, FORMULARYID, LIS_INDICATOR), потому что они варьировались в пределах одного и того же идентификатора. Я просто выбрал первый для каждого удостоверения личности, вы можете настроить его по своему вкусу.

...