def compute_features():
t = open("./package/training_set_3.csv", 'w')
c = csv.writer(t)
c.writerow(["serial no"])
i = 1
while i < 10:
c.writerow([i])
i = i + 1
t.close()
#two files 1_chat and 2_chat are created
#first feature to extract ////"""average words per line""""
average_word(
"./package/test_final.txt", 3
) #creates a text file named number_word_1 which contains number of word in each line.
#print "average_word_feature extracted"
#second feature """word length per line"''
word_length(
"./package/test_final.txt", 3
) #creates a text file named word_length_1 which contains number of word in each line.
#print "word_length_feature extracted"
#third feature ""uppercase_lowrcase ratio""""
ratio(
"./package/test_final.txt", 3
) #creates a text file named word_length_1 which contains number of word in each line.
#print "ratio_feature extracted"
#fourth feature ""time_chat , time to give that particular reply"""
#sixth feature """smiley count """every 10 line feature usage
smiley_count("./package/test_final.txt", 3)
#print "smiley count extracted"
#fifth feature """stopwaord usagee in a line""""every 10 line feature
stopword_usage("./package/test_final.txt", 3)
#print "stopword feature extracted"
#seventh feature """punchuation usage""""
punctuation("./package/test_final.txt", 3)
#print "punctuation_feature extracted"
#eighth feature """message length ""
msg_length("./package/test_final.txt", 3)
#print "msg_length_feature extracted"
#ninth feature """acronym count """
acro_line("./package/test_final.txt", 3)
def compute_features():
t = open("./package/training_set_3.csv", "w")
c = csv.writer(t)
c.writerow(["serial no"])
i = 1
while i < 10:
c.writerow([i])
i = i + 1
t.close()
# two files 1_chat and 2_chat are created
# first feature to extract ////"""average words per line""""
average_word(
"./package/test_final.txt", 3
) # creates a text file named number_word_1 which contains number of word in each line.
# print "average_word_feature extracted"
# second feature """word length per line"''
word_length(
"./package/test_final.txt", 3
) # creates a text file named word_length_1 which contains number of word in each line.
# print "word_length_feature extracted"
# third feature ""uppercase_lowrcase ratio""""
ratio(
"./package/test_final.txt", 3
) # creates a text file named word_length_1 which contains number of word in each line.
# print "ratio_feature extracted"
# fourth feature ""time_chat , time to give that particular reply"""
# sixth feature """smiley count """every 10 line feature usage
smiley_count("./package/test_final.txt", 3)
# print "smiley count extracted"
# fifth feature """stopwaord usagee in a line""""every 10 line feature
stopword_usage("./package/test_final.txt", 3)
# print "stopword feature extracted"
# seventh feature """punchuation usage""""
punctuation("./package/test_final.txt", 3)
# print "punctuation_feature extracted"
# eighth feature """message length ""
msg_length("./package/test_final.txt", 3)
# print "msg_length_feature extracted"
# ninth feature """acronym count """
acro_line("./package/test_final.txt", 3)
while i<5000:
c.writerow([i])
d.writerow([i])
c1.writerow([i])
d1.writerow([i])
#print i
i=i+1
t.close()
n.close()
t1.close()
n1.close()
#first feature to extract ////""average words per line""
average_word('./chats_process/'+str(first)+'_'+str(second)+'/'+first+'.txt',str(first),str(second),1)
#creates a text file named number_word_1 which contains number of word in each line inside the folder chats_process/user1_user2.
average_word('./chats_process/'+str(first)+'_'+str(second)+'/'+second+'.txt',str(first),str(second),2)
#print "average_word_feature extracted"
#second feature ""word length per line"''
word_length('./chats_process/'+str(first)+'_'+str(second)+'/'+first+'.txt',str(first),str(second),1)
#creates a text file named word_length_1 which contains number of word in each line.
word_length('./chats_process/'+str(first)+'_'+str(second)+'/'+second+'.txt',str(first),str(second),2)
#print "word_length_feature extracted"
#third feature ""uppercase_lowrcase ratio""
ratio('./chats_process/'+str(first)+'_'+str(second)+'/'+first+'.txt',str(first),str(second),1)
#creates a text file named word_length_1 which contains number of word in each line.
ratio('./chats_process/'+str(first)+'_'+str(second)+'/'+second+'.txt',str(first),str(second),2)
i=i+1
t.close()
n.close()
t1.close()
n1.close()
#f = open("package/"+file[1], 'r')
#seperating the two chats
print f
extract(f)
#two files 1_chat and 2_chat are created
#first feature to extract ////"""average words per line""""
average_word("./package/1_chat.txt",1)#creates a text file named number_word_1 which contains number of word in each line.
average_word("./package/2_chat.txt",2)
print "average_word_feature extracted"
#second feature """word length per line"''
word_length("./package/1_chat.txt",1)#creates a text file named word_length_1 which contains number of word in each line.
word_length("./package/2_chat.txt",2)
print "word_length_feature extracted"
#third feature ""uppercase_lowrcase ratio""""
ratio("./package/1_chat.txt",1)#creates a text file named word_length_1 which contains number of word in each line.
ratio("./package/2_chat.txt",2)
print "ratio_feature extracted"
#fourth feature ""time_chat , time to give that particular reply"""