gistfile1.txt

###################################################################
####################### DA: Assignment A7 #########################
########################################### Vivek Narang ##########

# Start #

# Importing required libraries #
library(readr)
library(stringr)
library(sentimentr)

# Reading the text file containing text from the Universal Declaration of Human Rights #
mystring <- read_file("C:\\A7.txt")

# Removing \r from string #
mystring2 <- str_replace_all(mystring, "\r", " ")

# Removing \n from string #
mystring3 <- str_replace_all(mystring2, "\n", " ")

# Removing 'Article' from string #
mystring4 <- str_replace_all(mystring3, "Article", "")

for(i in seq(0, 6, by = 2)) {
  
  for(j in seq(0, 4, by = 2)) {
    
    y <- sentiment(mystring4, n.before=i, n.after=j, amplifier.weight=1)
    write(paste("(n.before=", i, ", n.after=",j,"): ",mean(y$sentiment), sep="") , stdout())
  
    }
  
}


#####################################################################
# Inference:
# The mean of the sentiment value is positive across all combinations
# Increasing n.before is increasing means while increasing n.after
# is decreasing the mean values. The overall positive number from 
# sentiment analysis tells us that there is an overall positive tone 
# in the text. The n.before is a parameter which configures how many
# words to look for before the polarized word during the analysis. 
# The n.after parameter configures the number of words to use after
# the polarized word. With n.before increasing there is a general 
# increase in the mean sentiment values, however with the increase in
# n.after values the mean is decreasing for each n.before value.
#####################################################################

# End #