Maraton danych - eRka Kraków

maraton.r

library(dplyr)
data <- read.csv("gau12jbs Master 100715.csv", header=TRUE)

names <- select(data, starts_with("RV"), starts_with("RC"), starts_with("SW"), starts_with("SN"), starts_with("WC"), starts_with("AP"), starts_with("PH"), starts_with("SD"), starts_with("SY"), starts_with("RH"), starts_with("FS") ) %>%
  na.omit %>%
  row.names

train_names <- sample(names, length(names) * 7/10, replace = FALSE)
test_names <- setdiff(names, train_names)

train_set <- filter(data, row.names(data) %in% train_names) %>%
  select(starts_with("RV"), starts_with("RC"), starts_with("SW"), starts_with("SN"), starts_with("WC"), starts_with("AP"), starts_with("PH"), starts_with("SD"), starts_with("SY"), starts_with("RH"), starts_with("FS") ) %>%
  na.omit
test_set <- filter(data, row.names(data) %in% test_names) %>%
  select(starts_with("RV"), starts_with("RC"), starts_with("SW"), starts_with("SN"), starts_with("WC"), starts_with("AP"), starts_with("PH"), starts_with("SD"), starts_with("SY"), starts_with("RH"), starts_with("FS") ) %>%
  na.omit

train_set <- train_set[!sapply(train_set, is.logical)]
train_set <- cbind(filter(data, row.names(data) %in% train_names) %>% select(Language), train_set[!sapply(train_set, is.factor)])
test_set <- test_set[!sapply(test_set, is.logical)]
test_set <- cbind(filter(data, row.names(data) %in% test_names) %>% select(Language), test_set[!sapply(test_set, is.factor)])


summary(pc.cr <- princomp(train_set[,2:1040]))
loadings(pc.cr)

library(lattice)
pc.cr$scores
pca.plot <- xyplot(pc.cr$scores[,1] ~ pc.cr$scores[,2])
pca.plot$xlab <- "First Component"
pca.plot$ylab <- "Second Component"
pca.plot
biplot(pc.cr)