WARN: Some values are hard coded!!

Dimensionality Reduction from 7395 to 100

Is it reasonable ??

library(lsa)

print("lsa loaded")

TD <- read.csv(file="lyrics.spring.tdm",head=FALSE,sep=",")

print("csv loaded")

TD.svd=svd(TD)

print("svd finished")

rtd=TD.svd$u[,1:100] %*% diag(TD.svd$d[1:100]) %*% t(TD.svd$v[,1:100])

print("svd multiplication finished")

i=0;while(i < 7395){i=i+1; write(cosine(rtd[,1:1],rtd[,i:i]),file="lyrics.spring.csv", append=TRUE)}

print("all done")

http://www.behind-the-enemy-lines.com/2008/01/are-you-bayesian-or-frequentist-or.html

아 내가 그렇다는 것은 아니고 ㅎㅎ

https://en.wikipedia.org/wiki/Design_of_experiments

자동차관련 3 종세트, 워셔액

우유

AA

종이컵 큰거

방수팩

썬그라스