Various measurements corresponding to the same person are taken at different points of time resulting into DF1 and DF2 as shown below (this is only a sample of the hundreds of observations). Assuming the identity (ID) of the persons is not known (unlabeled classification), I want to match observation which belong to the same person. I thought of matching them using least Euclidean distance but provided very poor match. Any idea on how to do this in R, e.g., probability based matching?
ID <- letters[1:10]
lpl <- c(729,718,715,632,634,779,649,773,659,746)
wb1 <- c(153,173,113,130,158,176,178,165,126,193)
wb2 <- c(42,44,42,43,43,44,43,45,41,42)
wb3 <- c(330,308,356,300,295,349,292,328,274,330)
wb4 <- c(39,104,40,41,41,43,40,42,40,101)
DF1 <- data.frame(ID,lpl,wb1,wb2,wb3,wb4)
ID <- letters[1:10]
lpl <- c(703,681,672,599,574,712,590,707,621,721)
wb1 <- c(161,171,113,129,154,168,181,155,128,193)
wb2 <- c(41,44,41,42,41,42,41,41,41,41)
wb3 <- c(339,309,350,303,290,332,288,312,279,328)
wb4 <- c(39,105,38,38,41,40,39,39,41,99)
DF2 <- data.frame(ID,lpl,wb1,wb2,wb3,wb4)
