I write a function in R as following, but I could not use it in a vectorize way (the second part of the code below). I am wondering how can I make this function vectorize
normalize=function(browser){ if (browser=="Chrome" | browser=="Firefox" | browser=='Safari' | grepl('IE',browser)){ browser }else{ "Others" } } data$browser_n<-normalize(data$browser) // not working (data is a data frame) Use ifelse():
normalize=function(browser){ ifelse(browser=="Chrome" | browser=="Firefox" | browser=='Safari' | grepl('IE',browser), browser, 'Others') } browser <- c('aaa', 'Chrome') normalize(browser) # [1] "Others" "Chrome" 
normalize <- function(browser){ replace(browser, !(browser %in% c("Chrome", "Firefox", "Safari") | grepl('IE',browser)), "Others") } Here are two comments.
It would be better to use || instead of |. Here is why.
& and && indicate logical AND and | and || indicate logical OR. The shorter form performs elementwise comparisons in much the same way as arithmetic operators. The longer form evaluates left to right examining only the first element of each vector. Evaluation proceeds only until the result is determined. The longer form is appropriate for programming control-flow and typically preferred in if clauses.
Another approach would be to use any.
normalize = function(browser){ if (any(browser == "Chrome", browser == "Firefox", browser == "Safari", grepl("IE", browser)) { browser } else { "Others } } While this may not be most correct answer, you can try Vectorize. In general, you can use Vectorize on many functions to vectorize them. I should add that Vectorize is just a pretty wrapper around mapply
normalize = function(browser) { if (browser == "Chrome" | browser == "Firefox" | browser == "Safari" | grepl("IE", browser)) { return(browser) } else { return("Others") } } vNormalize <- Vectorize(normalize) data <- data.frame(browser = c("Chrome", "Firefox", "Safari", "IE 10")) vNormalize(data$browser) ## [1] Chrome Firefox Safari IE 10 ## Levels: Chrome Firefox IE 10 Safari 
Vectorize on your own functions is not generally good practiceOr, if you are not really interested in vectorization, but want to get the new variable browser_n as in your example, you could just write
data$browser_n<-data$browser data$browser_n[!(data$browser=="Chrome" | data$browser=="Firefox" | data$browser=='Safari' | grepl('IE',data$browser))] <- "Others" 