I have a huge data table with millions of rows that consists of Merchandise code with its description. I want to assign a category to each group (based on the combination of code and description). The problem is that the description is spelled in different ways and I want to convert all the similar names into a single one. Here is an illustrative example:
ibrary(data.table)
dt <- data.table(code = c(rep(1,2),rep(2,2),rep(3,2)), name = c('McDonalds','Mc
Dnald','Macys','macy','Comcast','Com-cats'))
dt[,cat:='NA']
setkeyv(dt,c('code','name'))
dt[.(1,'McDonalds'),cat:='Restaurant']
dt[.(1,'Mc Dnald'),cat:='Restaurant']
dt[.(1,'Macys'),cat:='Department Store']
Of course in the real case, it is impossible to go through all the spelling that refer to the same word and fix them manually. Is there a way to detect all the similar words and convert them to a single (correct) spelling?
Thanks in advance