Using Association rule mining in Semiconductor manufacturing

Association rule is an ideally method for collection type problem mining in semiconductor manufacturing. The trick is in setting the support and confidence level.
The initial setting maybe the ((100-baseline yield)/5)/100 for support, 0.7 for confidence.
support(A) = # of event A/ Total ==> If an equipment is malfunction, most of the "bad" must come from that equipment. We regards the low yields as "bad", if more than 1/5 of the bads occurred, it's an candidate.

Use scan to read data -- modified code

在做what的設定時 需要是list component.
也就是說當你設了what=list(...)之後 在讀資料時會以一個column一個column來讀, 每個column的屬性就是what所設定的
所以當你把what="character"時 其實是把所有的資料都當做是一個column的character
用例子來說明

example:
原始的資料檔如下
Isat_N43_mA_10/.32,7.1,6.2,5.3
Isat_N4_mA_10/.18,7.35,6.45,5.55
Isat_P43_mA_10/.27,-2.46,-2.91,-3.36
Isat_P4_mA_10/.18,-2.32,-2.71,-3.1

當用scan(infile, what="character",sep=",")
Read 16 items
[1] "Isat_N43_mA_10/.32" "7.1" "6.2" "5.3"
[5] "Isat_N4_mA_10/.18" "7.35" "6.45" "5.55"
[9] "Isat_P43_mA_10/.27" "-2.46" "-2.91" "-3.36"
[13] "Isat_P4_mA_10/.18" "-2.32" "-2.71" "-3.1"


當用scan(infile, what=list("character",bouble(0),double(0),double(0)),sep=",")
讀進去後的結果:

Read 4 records

[[1]]
[1] "Isat_N43_mA_10/.32" "Isat_N4_mA_10/.18" "Isat_P43_mA_10/.27" "Isat_P4_mA_10/.18"

[[2]]
[1] 7.10 7.35 -2.46 -2.32

[[3]]
[1] 6.20 6.45 -2.91 -2.71

[[4]]
[1] 5.30 5.55 -3.36 -3.10

要把資料轉成dataframe 只須用as.data.frame即可(data frame 其實就是vector of list 所以實在是不需要多此一舉的)

修改過的scan

	scanData.go <- function(inFileName, colName, colType, headerFg = FALSE, sep = ",", ...){
	if(length(colName)>length(colType[!colType == "NULL"])) stop("more column names than columns")
	skip <- 0
	if(headerFg){
	skip <- 1
	}
	tmp1 <- vector("list",length(colType))
	notNull <- which(!colType == "NULL", arr.ind = T)
	for(i in 1:length(notNull)){
	tmp <- colType[notNull[i]]
	switch(match(tmp,c("char","num","logical")),
	tmp1[[notNull[i]]] <- "character",
	tmp1[[notNull[i]]] <- double(0),
	tmp1[[notNull[i]]] <- logical)
	}
	out <- scan(inFileName, what = tmp1, skip = skip, sep = sep)
	out <- out[!(colType == "NULL")]
	if(length(colName) < length(out)){
	colName <- c(colName, paste("V", 1:(length(out)-length(colName)), sep = ""))
	warning("header and 'col.names' are of different lengths")
	}
	names(out) <- colName
	return(out)
	}

view raw scanData hosted with ❤ by GitHub