No missing values are allows kNN in R

Question

I've data set of 45212 elements with 17 columns and i want to find the class label of last column using kNN algorithm, according to me everything is OK, but I always come up with error

"Error in knn(train = data_train, test = data_test, cl = data_train_labels,  : 
  no missing values are allowed"

here is my code

> data_train <-data[1:25000,]
> data_test <-data[25001:45212,]
> data_train_labels <- data[1:25000, 17]
> data_test_labels <- data[1:25000, 17]
> install.package("class")
> library(class)
> data_test_pred <- knn(train=data_train, test=data_test, cl=data_train_labels, k=10)

here is how my data set looks like:

age,job,marital,education,default,balance,housing,loan,contact,day,month,duration,campaign,pdays,previous,poutcome,y
58,management,married,tertiary,no,2143,yes,no,unknown,5,may,261,1,-1,0,unknown,no
44,technician,single,secondary,no,29,yes,no,unknown,5,may,151,1,-1,0,unknown,no
33,entrepreneur,married,secondary,no,2,yes,yes,unknown,5,may,76,1,-1,0,unknown,no
47,blue-collar,married,unknown,no,1506,yes,no,unknown,5,may,92,1,-1,0,unknown,no
33,unknown,single,unknown,no,1,no,no,unknown,5,may,198,1,-1,0,unknown,no
35,management,married,tertiary,no,231,yes,no,unknown,5,may,139,1,-1,0,unknown,no
28,management,single,tertiary,no,447,yes,yes,unknown,5,may,217,1,-1,0,unknown,no
42,entrepreneur,divorced,tertiary,yes,2,yes,no,unknown,5,may,380,1,-1,0,unknown,no
58,retired,married,primary,no,121,yes,no,unknown,5,may,50,1,-1,0,unknown,no
43,technician,single,secondary,no,593,yes,no,unknown,5,may,55,1,-1,0,unknown,no
41,admin.,divorced,secondary,no,270,yes,no,unknown,5,may,222,1,-1,0,unknown,no

Answer 1

I think that your problem is all of the factors in your data. The knn documentation says that it uses Euclidean distance, which does not make sense for factors. Here is a possible solution if you really want to use knn. You can get a distance matrix between the points using daisy in the cluster package. There are several implementations of knn in R but I don't know of one that accepts a distance matrix. You could either write your own (not so difficult) or you could map the distance matrix to a Euclidean space using cmdscale. Then use knn on the projected space.

Answer 2

I believe that your mistake is: data_train <-data[1:25000,]

You are including your header that you have not normalized. I was able to reproduce the same error. But when I changed to data_train <-data[2:25000,] it ran fine.