I'm running it in the cloud environment, but I tried running it on my desktop as well. It works in the cloud environment with a 1500 line dataset.
Run this search by the algorithm for CNN Data
search_res_CNN <- merge(search_dtCNN[, id:=1L], search_for[, id:=1L], by="id", allow.cartesian=TRUE)[,
match:=corpusCNN %like% word, by=.(corpusCNN, word, value)][
match==TRUE, .(words=paste(sort(word), collapse=", "), "CNN Insider Word Score"=sum(value)), by=corpusCNN]
search_res_CNN <- merge(search_dtCNN[, -"id"], search_res_CNN, on="corpusCNN", all.x=TRUE)