def testKnownTFIDF(self):
"""
Testing to see whether the tfidf values for arbitrarily selected words
in the articles correspond with manually calculated values.
"""
articleList = []
theList = []
for string in self.strings:
articleList.append(tfidf.tf(string))
for string in self.theTwentyFive:
theList.append(tfidf.tf(string))
idfArtDict = tfidf.idf(articleList)
idfTheDict = tfidf.idf(theList)
tfidfArtList = tfidf.tfidf(idfArtDict, articleList)
tfidfTheList = tfidf.tfidf(idfTheDict, theList)
self.assertEqual(tfidfArtList[1]["Meditation"], math.log10(6/1) * (1/19))
self.assertEqual(tfidfArtList[2]["books"], math.log10(6/1) * (1/18))
self.assertEqual(tfidfArtList[5]["the"], math.log10(6/3) * (5/5))
self.assertEqual(tfidfTheList[3]["the"], math.log10(5/5) * (5/5))
def weight(d):
global idf, x, sq
total_words = sum(d.values())
dsq = 0
d1 = {}
for word in d:
if word in idf:
d1[word] = tfidf.tf(d[word], total_words) * idf[word]
dsq += d1[word] * d1[word]
return d1, dsq
def testKnownTF(self):
"""
Testing to see if the term frequencies for words match up with manual
tf calculations.
"""
tfDict = tfidf.tf(self.string1)
self.assertEqual(tfDict["meditation"], (1/19))
tfDict = tfidf.tf(self.string2)
self.assertEqual(tfDict["be"], (3/18))
tfDict = tfidf.tf(self.string3)
self.assertEqual(tfDict["dog"], (0/11))
tfDict = tfidf.tf(self.string4)
self.assertEqual(tfDict["bureaucracy."], (1/12))
tfDict = tfidf.tf(self.string5)
self.assertEqual(tfDict["the"], (5/5))
def testArticleOrder(self):
"""
Testing to see whether the articles in the articleList retain their
order. This is the order that they will be in for testKnownTFIDF.
"""
articleList = []
for string in self.strings:
articleList.append(tfidf.tf(string))
self.assertEqual(articleList[1]["Meditation"], (1/19))
self.assertEqual(articleList[2]["be"], (3/18))
self.assertEqual(articleList[3]["can't"], (1/11))
self.assertEqual(articleList[4]["bureaucracy."], (1/12))
self.assertEqual(articleList[5]["the"], (5/5))
def chapterSummary(chap):
unitSummary = []
for unit in chap:
summarizedText = summarization.generate_summary(unit, 2)
if summarizedText == []:
continue
tfidfText = tfidf.tf(unit)
sentence = ''
for x in summarizedText:
sentence += x
print(tfidfText, '------> ', sentence)
entry = {"query": tfidfText, "ans": sentence}
insert = collection.insert_one(entry)
print(insert.inserted_id)
unitSummary.append(summarizedText)
return unitSummary
def testLength(self):
"""
Testing the lengths of the dictionaries to see if they hold an
accurate count of unique words.
"""
tfDict = tfidf.tf(self.emptyString)
self.assertEqual(len(tfDict), 0)
tfDict = tfidf.tf(self.string1)
self.assertEqual(len(tfDict), 17)
tfDict = tfidf.tf(self.string2)
self.assertEqual(len(tfDict), 13)
tfDict = tfidf.tf(self.string3)
self.assertEqual(len(tfDict), 11)
tfDict = tfidf.tf(self.string4)
self.assertEqual(len(tfDict), 10)
tfDict = tfidf.tf(self.string5)
self.assertEqual(len(tfDict), 1)
print "TFIDF for knn"
print tfidf.fast_tf_idf(files, "knn")
print "TFIDF for neural"
print tfidf.fast_tf_idf(files, "neural")
print "TFIDF for network"
print tfidf.fast_tf_idf(files, "network")
print "TFIDF for deep"
print tfidf.fast_tf_idf(files, "deep")
# Classification by TF
Y = ["learning", "knn", "neural", "network", "deep"]
X = []
for f in files:
temp = []
for word in Y:
temp.append(tfidf.tf(f, word))
X.append(temp)
print "TF for [learning, knn, neural, network, deep]"
print X
# Hamming Distance
X = []
for f1 in files:
Y = []
for f2 in files:
Y.append(distance.hammingDistance(f1, f2))
X.append(Y)
print "Matrix of Hamming Distance between all files"
plt.matshow(X)
plt.show()
# import fungsi fari tfidf
from tfidf import tf
from tfidf import idf
# variable
n_term = 3
total_term = 100
n_docs = 10000000
total_docs = 1000
# memanggil fungsi tf untuk menghitung term frequency
# variabel tf_value akan menampung file dari hasil komparasi fungsi tf
tf_value = tf(n_term, total_term)
idf_value = idf(n_docs, total_docs)
# print tf_value
print("Term frequency : {0}".format(tf_value))
print("IDF : {0}".format(idf_value))
# Bobot
bobot = tf_value * idf_value
print("Weight : {0}".format(bobot))
#import fungsi dari file tfidf
from tfidf import tf, idf
#variable
n_terms = 3
total_terms = 100
n_docs = 10000000
n_docs_with_term = 1000
#memanggil fungsi tf untuk menghitung term frequency
#variable tf_value akan menampung file dari hasil komputasi fungsi tf
tf_value = tf(n_terms, total_terms)
idf_value = idf(n_docs, n_docs_with_term)
#print tf_value
print("Term frequency: {0}".format(tf_value))
print("Inverse document frequency: {0}".format(idf_value))
tfidf_value = tf_value * idf_value
print("Tf * idf: {0}".format(tfidf_value))
# -*- coding: utf-8 -*-
# @Time : 2017/5/21 下午 04:30
# @Author : Yuhsuan
# @File : test.py
# @Software: PyCharm Community Edition
from nltk.corpus import reuters
fileid = 'training/3386'
print(" ".join(reuters.words(fileids=fileid)))
print(reuters.categories(fileids=fileid))
import tfidf as ti
print(ti.tf('Portland', fileid))
