def beta_money_Detection(txt):
beta_value = betafunc.Beta(txt)
#print(beta_value)
upper_limit = 0.19
lower_limit = 0.057
std = 0.025
if(beta_value + std*2.5 < upper_limit):
return 5.0
elif(beta_value + std <= upper_limit or beta_value - std >= lower_limit):
return 2.5
else:
return 0.0
def beta_for_imgntn(text):
beta_value = betafunc.Beta(text)
lower_limit = 0.48
upper_limit = 1.25
std = 0.17
if(beta_value > 0.8):
return 5.0
elif(beta_value-std > lower_limit or beta_value+std < upper_limit):
#print(beta_value)
return 2.5
else:
return 0.0
def betaPublicisticDetection(text):
grade = 0.0
beta = betafunc.Beta(text)
sent_len = average_length_of_word.average_sent_length(text)
prefix = ScienceDetectionFuncs.numPrefixCheck(text)
text_vector = tags_counter.counter(text)
if (text_vector[0] < 0.025 and beta < 0.5):
grade = 2.5
if (prefix < 70):
grade = 5.0
return grade
else:
return grade
else:
return grade
def publicisticDetection(text):
grade = 0.0
#vector = tags_counter.counter(text)
beta = betafunc.Beta(text)
prefix = ScienceDetectionFuncs.numPrefixCheck(text)
NN_value = same_neighbors_counter.NN_same_neighbors(text)
if (beta > 0.22 and prefix < 90):
"""Сейчас этот текст скорее всего не имеет признаки делового или научного стилей"""
grade = 2.5
#print(NN_value)
if (NN_value >= 400):
"""Сейчас этот текст имеет признаки публицистического текста"""
grade = 5.0
#print(grade)
return grade
else:
#print(grade)
return grade
else:
#print(grade)
return grade
def scienceParamDetection(txt):
grade = 0
prefix = numPrefixCheck(txt)
tags_vector = tags_counter.counter(txt)
#sent_len = average_length_of_word.average_sent_length(txt)
beta = betafunc.Beta(txt)
#print(prefix, beta)
if (prefix > 100 and beta > 0.22):
grade = 2.5
#print("Данный текст почти невозможно переделать в текст делового или публицистического стиля \n")
if (tags_vector[5] < 1936908127739503 / 4611686018427387904
or tags_vector[4] < 5404319552844595 / 72057594037927936):
grade = 5
#print("Есть признаки схожести с научным текстом")
return grade
else:
#print("Возможно схоже с художестным текстом")
return grade
return grade
else:
#print("There is nothing about science here")
return 0
maxFilename = ["", "", "", "", "", "", ""]
average_prefix = [0, 0, 0, 0]
average_m_prefix = [0, 0, 0, 0]
for file in iner_list_of_dir:
if (file.endswith("utf.txt")):
with open(folder + "/" + file, "r", encoding="utf-8") as currfile:
text = currfile.read()
freq = tags_counter.counter(text)
average_m_prefix[
itr] += MoneyDetectionFuncs.anti_mutex_prefix_Detection(
text)
average_prefix[itr] += ScienceDetectionFuncs.numPrefixCheck(
text)
#std_list.append(average_prefix[itr])
amount_of_words[itr] += len(nltk.tokenize.word_tokenize(text))
beta += betafunc.Beta(text)
average_word_length += average_length_of_word.average_word_length(
text)
average_sent_length += average_length_of_word.average_sent_length(
text)
itr2 = 0
for feature in maximus:
if (maximus[itr2] < freq[itr2]):
maximus[itr2] = freq[itr2]
maxFilename[itr2] = file
itr2 += 1
print(average_prefix[itr])
print(average_m_prefix[itr])
#print(beta)
print(freq)
freqall[itr] = [x + y for x, y in zip(freq, freqall[itr])]