R help - Apr 2018 - Check if row of dataframe is superset of any row in another dataframe.

Hi,

I am looking for a way in which I can check if rows in 1 dataframe are
present in another data frame in a unique way. A row in dataframe should be
super set of any row in another dataframe.

I can write a for loop for it, however, that will be inefficient. So, I am
looking for an efficient way to do this in R.

I have explained it with an example below:

I want to check if a row in dataframe B is:
1) either equal to any row in A or
2) has 1's atleast for the columns where (any) row in B has 1's.

My output/result is a vector of 1(TRUE) or 0(FALSE) of length equal to
number of rows in B. The first row in B is exactly present in A so result
has first bit as 1. Second row in B has matches with 2nd row of dataframe A
(it has an extra 1 in 3rd column,which is ok);so second bit of result is
also 1. Similarly, the 3rd row of B, can match to any row in A, so 3rd bit
in result is also a 1. Next, 4th row in B has 1 for a column where no row
in A has 1, so last bit in result is 0.

Dataframe A
1 0 1 0
1 1 0 0
0 1 1 0

Dataframe B
1 0 1 0
1 1 1 0
1 1 1 1
0 0 0 1

Result<- 1 1 1 0

Thanks for the help,
Neha

	[[alternative HTML version deleted]]

Hi Neha,
How about this?

A <- as.matrix(A)
B <- as.matrix(B)

C   <- A %*% t(B)
SA  <- apply(A, MAR=1, sum )
SB  <- apply(B, MAR=1, sum )
vapply( 1:nrow(B), function(j) { sum( C[,j]==SA & SA <= SB[j] ) > 0 },
1 )

HTH,
Eric


On Sat, Apr 21, 2018 at 10:27 AM, Neha Aggarwal <aggarwalneha2000 at
gmail.com>
wrote:
> Hi,
>
> I am looking for a way in which I can check if rows in 1 dataframe are
> present in another data frame in a unique way. A row in dataframe should be
> super set of any row in another dataframe.
>
> I can write a for loop for it, however, that will be inefficient. So, I am
> looking for an efficient way to do this in R.
>
> I have explained it with an example below:
>
> I want to check if a row in dataframe B is:
> 1) either equal to any row in A or
> 2) has 1's atleast for the columns where (any) row in B has 1's.
>
> My output/result is a vector of 1(TRUE) or 0(FALSE) of length equal to
> number of rows in B. The first row in B is exactly present in A so result
> has first bit as 1. Second row in B has matches with 2nd row of dataframe A
> (it has an extra 1 in 3rd column,which is ok);so second bit of result is
> also 1. Similarly, the 3rd row of B, can match to any row in A, so 3rd bit
> in result is also a 1. Next, 4th row in B has 1 for a column where no row
> in A has 1, so last bit in result is 0.
>
> Dataframe A
> 1 0 1 0
> 1 1 0 0
> 0 1 1 0
>
> Dataframe B
> 1 0 1 0
> 1 1 1 0
> 1 1 1 1
> 0 0 0 1
>
> Result<- 1 1 1 0
>
> Thanks for the help,
> Neha
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/
> posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
	[[alternative HTML version deleted]]

Hi Neha,
How about this?

find_subset<-function(x,y) {
 yrows<-dim(y)[1]
 match<-0
 for(row in 1:yrows) match<-sum(x&y[row]) >= sum(y[row])
 return(match)
}
apply(B,1,find_subset,A)

This is somewhat obscure, as the dataframe B is coerced to a matrix by
the apply function.

Jim

On Sat, Apr 21, 2018 at 5:27 PM, Neha Aggarwal
<aggarwalneha2000 at gmail.com> wrote:
> Hi,
>
> I am looking for a way in which I can check if rows in 1 dataframe are
> present in another data frame in a unique way. A row in dataframe should be
> super set of any row in another dataframe.
>
> I can write a for loop for it, however, that will be inefficient. So, I am
> looking for an efficient way to do this in R.
>
> I have explained it with an example below:
>
> I want to check if a row in dataframe B is:
> 1) either equal to any row in A or
> 2) has 1's atleast for the columns where (any) row in B has 1's.
>
> My output/result is a vector of 1(TRUE) or 0(FALSE) of length equal to
> number of rows in B. The first row in B is exactly present in A so result
> has first bit as 1. Second row in B has matches with 2nd row of dataframe A
> (it has an extra 1 in 3rd column,which is ok);so second bit of result is
> also 1. Similarly, the 3rd row of B, can match to any row in A, so 3rd bit
> in result is also a 1. Next, 4th row in B has 1 for a column where no row
> in A has 1, so last bit in result is 0.
>
> Dataframe A
> 1 0 1 0
> 1 1 0 0
> 0 1 1 0
>
> Dataframe B
> 1 0 1 0
> 1 1 1 0
> 1 1 1 1
> 0 0 0 1
>
> Result<- 1 1 1 0
>
> Thanks for the help,
> Neha
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.