def feature_seven(text):
words = get_words(text)
syllables_sum = 0
for word in words:
syllables_sum += count_syllables(word)
return 0.39 * len(words)/total_sentences(text) + 11.8 * syllables_sum/len(words) - 15.59
def feature_three(text):
words = get_words(text)
poly_syllables_words = 0
for word in words:
if count_syllables(word) > 1:
poly_syllables_words += 1
return poly_syllables_words / total_sentences(text)
def feature_three(text):
words = get_words(text)
poly_syllables_words = 0
for word in words:
if count_syllables(word) > 1:
poly_syllables_words += 1
return poly_syllables_words/total_sentences(text)
def feature_seven(text):
words = get_words(text)
syllables_sum = 0
for word in words:
syllables_sum += count_syllables(word)
return 0.39 * len(words) / total_sentences(
text) + 11.8 * syllables_sum / len(words) - 15.59
def feature_five(text):
path_difficult_words = "/Users/Ivan/PycharmProject/ReadAbility/DataSets_raw/DaleChallEasyWordList.txt"
words = get_words(text)
difficult_words_sum = 0
with open(path_difficult_words, 'r') as f:
difficult_words = f.readlines()
for word in words:
if word not in difficult_words:
difficult_words_sum += 1
return 0.0496 * len(words)/total_sentences(text) + 0.1579 * difficult_words_sum/len(words) * 100 + 3.6365
def extract_features(data):
root = ElementTree.fromstring(data)
pure_text = root[0].text
ne = extract_entities_api(data)
tw = len(get_words(pure_text))
ts = total_sentences(pure_text)
feature1 = ne/tw*100
feature2 = ne/ts*100
return [feature1, feature2]
def extract_features(data):
root = ElementTree.fromstring(data)
pure_text = root[0].text
ne = extract_entities_api(data)
tw = len(get_words(pure_text))
ts = total_sentences(pure_text)
feature1 = ne / tw * 100
feature2 = ne / ts * 100
return [feature1, feature2]
def extract_features(data):
extr_entities = extract_entities(data)
ne = extr_entities.ne
tw = len(get_words(data))
ts = total_sentences(data)
feature1 = ne/tw*100
feature2 = ne/ts*100
print str(feature1) + " " + str(feature2)
return [feature1, feature2]
def extract_features(data):
extr_entities = extract_entities(data)
ne = extr_entities.ne
tw = len(get_words(data))
ts = total_sentences(data)
feature1 = ne / tw * 100
feature2 = ne / ts * 100
print str(feature1) + " " + str(feature2)
return [feature1, feature2]
def feature_five(text):
path_difficult_words = "/Users/Ivan/PycharmProject/ReadAbility/DataSets_raw/DaleChallEasyWordList.txt"
words = get_words(text)
difficult_words_sum = 0
with open(path_difficult_words, 'r') as f:
difficult_words = f.readlines()
for word in words:
if word not in difficult_words:
difficult_words_sum += 1
return 0.0496 * len(words) / total_sentences(
text) + 0.1579 * difficult_words_sum / len(words) * 100 + 3.6365
def extract_features(data, pos):
extr_words = extract_words(data, pos)
words = extr_words.words
unique_words = extr_words.unique_words
total_w = len(get_words(data))
total_unique_w = len(np.unique(get_words(data)))
total_s = total_sentences(data)
feature1 = words/total_w*100
feature2 = unique_words/total_w*100
feature3 = unique_words/total_unique_w
feature4 = words/total_s
feature5 = unique_words/total_s
return [feature1, feature2, feature3, feature4, feature5]
def extract_features(data, pos):
extr_words = extract_words(data, pos)
words = extr_words.words
unique_words = extr_words.unique_words
total_w = len(get_words(data))
total_unique_w = len(np.unique(get_words(data)))
total_s = total_sentences(data)
feature1 = words / total_w * 100
feature2 = unique_words / total_w * 100
feature3 = unique_words / total_unique_w
feature4 = words / total_s
feature5 = unique_words / total_s
return [feature1, feature2, feature3, feature4, feature5]
def extract_features(data, pos_type):
root = ElementTree.fromstring(data)
pure_text = root[0].text
extr_words = extract_words(data, pos_type)
words = extr_words.words
unique_words = extr_words.unique_words
total_w = len(get_words(pure_text))
total_unique_w = len(np.unique(get_words(pure_text)))
total_s = total_sentences(pure_text)
feature1 = words/total_w*100
feature2 = unique_words/total_w*100
feature3 = unique_words/total_unique_w
feature4 = words/total_s
feature5 = unique_words/total_s
return [feature1, feature2, feature3, feature4, feature5]
def extract_features(data, pos_type):
root = ElementTree.fromstring(data)
pure_text = root[0].text
extr_words = extract_words(data, pos_type)
words = extr_words.words
unique_words = extr_words.unique_words
total_w = len(get_words(pure_text))
total_unique_w = len(np.unique(get_words(pure_text)))
total_s = total_sentences(pure_text)
feature1 = words / total_w * 100
feature2 = unique_words / total_w * 100
feature3 = unique_words / total_unique_w
feature4 = words / total_s
feature5 = unique_words / total_s
return [feature1, feature2, feature3, feature4, feature5]
def feature_six(text):
return len(get_words(text))/total_sentences(text)
def feature_six(text):
return len(get_words(text)) / total_sentences(text)
