I'm making some text-mining from a corpus of words, and I'm having a textfile output with 3000 lines like this :
dns 11 11 [2, 355, 706, 1063, 3139, 3219, 3471, 3472, 3473, 4384, 4444]
xhtml 8 11 [1651, 2208, 2815, 3487, 3517, 4480, 4481, 4504]
javascript 18 18 [49, 50, 175, 176, 355, 706, 1063, 1502, 1651, 2208, 2280, 2815, 3297, 4068, 4236, 4480, 4481, 4504]
There is the word, the number of lines where it've appeared, the number of total appearances, and the n° of these lines.
I'm trying to calculate The Chi-squared Value, and that textfile is the input for my code below :
measure = nltk.collocations.BigramAssocMeasures()
dicto = {}
for i in lines :
tokens = nltk.wordpunct_tokenize(i)
m = tokens[0] #m is the word
list_i = tokens[4:]
list_i.pop()
for x in list_i :
if x ==',':
ind = list_i.index(x)
list_i.pop(ind)
dicto[m]=list_i #for each word i create a dictionnary with the n° of lines
#for each word I calculate the Chi-squared with every other word
#and my problem is starting right here i think
#The "for" loop and the z = .....
for word1 in dicto :
x=dicto[word1]
vector = []
for word2 in dicto :
y=dicto[word2]
z=[val for val in x if val in y]
#Contingency Matrix
m11 = cpt-(len(x)+len(y)-len(z))
m12 = len(x)-len(z)
m21 = len(y)-len(z)
m22 = len(z)
n_ii =m11
n_ix =m11+m21
n_xi =m11+m12
n_xx =m11+m12+m21+m22
Chi_squared = measure.chi_sq(n_ii, (n_ix, n_xi), n_xx)
#I compare with the minimum value to check independancy between words
if Chi_squared >3.841 :
vector.append([word1, word2 , round(Chi_square,3))
#The correlations calculated
#I sort my vector in a descending way
final=sorted(vector, key=lambda vector: vector[2],reverse = True)
print word1
#I take the 4 best scores
for i in final[:4]:
print i,
My problem is that the calcul is taking to much time (I'm talking about Hours !!) Is there anything that I can change ? anything that I do to improve my code ? Any other Python structures ? any ideas ?
There are a few opportunities for speedup, but my first concern is vector . Where is it initialized? In the code posted, it gets n^2 entries and sorted n times! That seems unintentional. Should it be cleared? Should final be outside the loop?
final=sorted(vector, key=lambda vector: vector[2],reverse = True)
is functional, but has ugly scoping, better is:
final=sorted(vector, key=lambda entry: entry[2], reverse=True)
In general, to solve timing issues consider using a profiler .
First, if for every word you have unique line numbers, use sets instead of lists: finding set intersection is much faster than the intersection of lists (especially if lists are not ordered).
Second, precompute list lengths - now you compute it twice for every single inner cycle step.
And third - use numpy
for this kind of computation.
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.