tr -d "[0-9]*[a-zA-z]*" < data_file.txt >newdata1.txt

tr -d "[-|.|(|)|/]" < newdata1.txt > newdata2.txt

tr -d "[-|,|'|\t|?|\"]" < newdata2.txt > newdata3.txt


# Command for counting frequency of word in the training set.
tr "[ ]*" "[\n*]" < new_Blog.txt | sort | uniq -c | sort -nr | sed -e 's/^[ \t]*//'  > new_Blog_freq.txt 


# Command for deleting frequency and '+' sign from the frequency file and make the data as a 2 dimensional array.
tr -d "[0-9] " < test_set.txt | tr "=|+|:" "[\t*]"  > new4.tx


for leading space remove 
sed "s/^[ \t]*//" -i output.txt > out.txt
