Btukfyl

Consider the following vector ‘tels’ which contains telephone numbers from “KANSAS”, “TEXAS” and “NEW YORK” regions.

tels <- c("510-548-2238", "707-231-2440", "650-752-1300", "510-674-3482", "510-853-5695", "510-882-9898", "650-555-6311", "707-885-6351", "650-231-1234", "650-096-0023", "707-691-6763")

If the number starts with 510, the phone number is from “KANSAS”, if it is 707, then “NEW YORK” and if it is 650 then the number is from “TEXAS”



Use R concepts and obtain the following dataframe as ouput.



               Expected Output:



                  PhoneNumbers            State



               1     5105482238       KANSAS



               2     7072312440   NEW YORK



               3     6507521300          TEXAS



               4     5106743482       KANSAS



               5     5108535695       KANSAS



               6     5108829898       KANSAS



               7     6505556311         TEXAS



               8     7078856351   NEW YORK



               9     6502311234         TEXAS



               10   6500960023         TEXAS



               11   7076916763   NEW YORK







This is my code :



z<-substr(tels,1,3)

dirt<-data.frame(tels,z)

dirt

for(i in z){

  if(i==510){

    sta<-"ddfdd"

  }if(i==707){

    sta<-"NEW YORK"

  }

  if((i==650)){

    sta<-"TEXAS"

  }



}

das<-data.frame(tels,sta)

das





but I'm getting this output:

          tels      sta

1  510-548-2238 NEW YORK

2  707-231-2440 NEW YORK

3  650-752-1300 NEW YORK

4  510-674-3482 NEW YORK

5  510-853-5695 NEW YORK

6  510-882-9898 NEW YORK

7  650-555-6311 NEW YORK

8  707-885-6351 NEW YORK

9  650-231-1234 NEW YORK

10 650-096-0023 NEW YORK

11 707-691-6763 NEW YORK

You can use factor with the labels being the state and labels being the first 3 digits

data.frame(tels,

 state = factor(substr(tels,0,3), c('510','650','707'), c('KANSAS','TEXAS','NEW YORK')))

           tels    state

1  510-548-2238   KANSAS

2  707-231-2440 NEW YORK

3  650-752-1300    TEXAS

4  510-674-3482   KANSAS

5  510-853-5695   KANSAS

6  510-882-9898   KANSAS

7  650-555-6311    TEXAS

8  707-885-6351 NEW YORK

9  650-231-1234    TEXAS

10 650-096-0023    TEXAS

11 707-691-6763 NEW YORK

We substr the 'tels' and then create a named vector to match the substr values and replace it with the values in the named vector

data.frame(PhoneNumbers = tels, state = setNames(c("KANSAS", "NEW YORK", "TEXAS"),

               c('510', '707', '650'))[substr(tels, 1, 3)])

#   PhoneNumbers    state

#1  510-548-2238   KANSAS

#2  707-231-2440 NEW YORK

#3  650-752-1300    TEXAS

#4  510-674-3482   KANSAS

#5  510-853-5695   KANSAS

#6  510-882-9898   KANSAS

#7  650-555-6311    TEXAS

#8  707-885-6351 NEW YORK

#9  650-231-1234    TEXAS

#10 650-096-0023    TEXAS

#11 707-691-6763 NEW YORK

You can find first pattern using ^510, ^650, and ^707. To easily add new column, I have used dplyr package.

library(tidyverse) # has dplyr and stringr

# data set -------------------------------

(dirt <- data_frame(PhoneNumbers = c("510-548-2238", "707-231-2440", "650-752-1300", "510-674-3482", "510-853-5695", "510-882-9898", "650-555-6311", "707-885-6351", "650-231-1234", "650-096-0023", "707-691-6763")))

#> # A tibble: 11 x 1

#>    PhoneNumbers

#>    <chr>       

#>  1 510-548-2238

#>  2 707-231-2440

#>  3 650-752-1300

#>  4 510-674-3482

#>  5 510-853-5695

#>  6 510-882-9898

#>  7 650-555-6311

#>  8 707-885-6351

#>  9 650-231-1234

#> 10 650-096-0023

#> 11 707-691-6763

You can make function to find each region by finding each pattern: stringr::str_detect()

You can do it at once using sapply(). If you perform str_detect to c("^510", "^650", "^707"), you will get a matrix each of which column is the number. Each value is whether the number contains the pattern(TRUE or FALSE), i.e. 11 x 3.

For each row, you have only one TRUE by construction. You can find this index and subset c("KANSAS", "TEXAS", "NEW YORK").

find_region <- function(x) {

  sta <- c("^510", "^650", "^707")

  stt <- sapply(sta, function(p) {

    str_detect(x, pattern = p)

  }) %>% # produce matrix 11x3 of TRUE and FALSE, each column = 510, 650, 707, TRUE if x contains the pattern

    apply(1, which) # get the index

  c("KANSAS", "TEXAS", "NEW YORK")[stt]

}

Using this function, you can add new column: dplyr::mutate()

dirt %>% 

  mutate(State = find_region(PhoneNumbers))

#> # A tibble: 11 x 2

#>    PhoneNumbers State   

#>    <chr>        <chr>   

#>  1 510-548-2238 KANSAS  

#>  2 707-231-2440 NEW YORK

#>  3 650-752-1300 TEXAS   

#>  4 510-674-3482 KANSAS  

#>  5 510-853-5695 KANSAS  

#>  6 510-882-9898 KANSAS  

#>  7 650-555-6311 TEXAS   

#>  8 707-885-6351 NEW YORK

#>  9 650-231-1234 TEXAS   

#> 10 650-096-0023 TEXAS   

#> 11 707-691-6763 NEW YORK