CODEWITHSUNDEEP
r
user3476463
user3476463

Reputation: 4575

Create New Fields from List Values

I have a dataframe of the form below which has two fields. field1 is a categorical field with only two values TRUE or FALSE, and field2 is a list. I would like to parse field2 and create new columns in my dataframe one for each unique value in the character list. For example, I would like to create 3 new columns from the data below, the columns would be Bas, ants, and onal. I would like the new columns to contain a flag TRUE or FALSE indicating that the field2 value for that row contained the value the field is named for. For example, the row 1 value for the new column Bas would be TRUE, the row 1 value of column onal would be FALSE. In Python there's a function called getdummies that does something similar. I'm not sure if there's an equivalent in r, also I'm not exactly sure how to parse the list. Any tips are greatly appreciated.

Sample data: 

structure(list(field1 = c("False", "TRUE"), field2 = list(
    c("Bas", "ants"), c("Bas", "onal"))), .Names = c("field1", 
"field2"), row.names = c(1904L, 1968L), class = "data.frame")

Here is a sample of what I'd like the output to look like:

structure(list(field1 = c(FALSE, TRUE), field2 = list(
    c("Bas", "ants"), c("Bas", "onal")), class = "factor"), Bas = c(TRUE, TRUE
), ants = c(TRUE, FALSE), onal = c(FALSE, TRUE)), .Names = c("field1", 
"field2", "Bas", "ants", "onal"), class = "data.frame", row.names = c(NA, 
-2L))

Upvotes: 0

Views: 80

Answers (2)

Pankaj Sharma
Pankaj Sharma

Reputation: 388

Try this.

data<-structure(list(field1 = c("False", "TRUE"), field2 = c("Bas", "Bas"),field3=c("ants", "onal")), .Names = c("field1", 
"field2","field3"), row.names = c(1904L, 1968L), class = "data.frame")

library(reshape2)    
newdata <-melt(data,id.vars=c("field3","field2"))    
x<-acast(newdata, value~field3)    
y<-acast(newdata, value~field2)    
final=cbind(x,y)

Upvotes: 0

akrun
akrun

Reputation: 887213

We can use mtabulate from library(qdapTools)

library(qdapTools)
!!(mtabulate(df1$field2)) 
#      ants  Bas  onal
#[1,]  TRUE TRUE FALSE
#[2,] FALSE TRUE  TRUE

data

df1 <- structure(list(field1 = c("False", "TRUE"),
field2 = list(c("Bas", 
"ants"), c("Bas", "onal"))), .Names = c("field1", "field2"),
row.names = c(1904L, 1968L), class = "data.frame")

Upvotes: 1

Related Questions