def lex_readability(self, text, mode='fre'):
if mode == 'all':
fre_score = textstat.flesch_reading_ease(text)
fog_index = textstat.gunning_fog(text)
fkg_index = textstat.flesch_kincaid_grade(text)
dcr_score = textstat.dale_chall_readability_score(text)
text_standard = textstat.text_standard(text, float_output=True)
return fre_score, fog_index, fkg_index, dcr_score, text_standard
if mode == 'fre':
fre_score = textstat.flesch_reading_ease(text)
return fre_score
if mode == 'fog':
fog_index = textstat.gunning_fog(text)
return fog_index
if mode == 'fkg':
fkg_index = textstat.flesch_kincaid_grade(text)
return fkg_index
if mode == 'dcr':
dcr_score = textstat.dale_chall_readability_score(text)
return dcr_score
if mode == 'text_std':
text_standard = textstat.text_standard(text, float_output=True)
return text_standard
def get_readibility(text, metric="flesch_kincaid_grade"):
"""
Return a score which reveals a piece of text's readability level.
Reference: https://chartbeat-labs.github.io/textacy/getting_started/quickstart.html
https://en.wikipedia.org/wiki/Flesch%E2%80%93Kincaid_readability_tests
"""
if metric == "flesch_kincaid_grade":
result = textstat.flesch_kincaid_grade(text)
elif metric == "flesch_reading_ease":
result = textstat.flesch_reading_ease(text)
elif metric == "smog_index":
result = textstat.smog_index(text)
elif metric == "coleman_liau_index":
result = textstat.coleman_liau_index(text)
elif metric == "automated_readability_index":
result = textstat.automated_readability_index(text)
elif metric == "dale_chall_readability_score":
result = textstat.dale_chall_readability_score(text)
elif metric == "difficult_words":
result = textstat.difficult_words(text)
elif metric == "linsear_write_formula":
result = textstat.linsear_write_formula(text)
elif metric == "gunning_fog":
result = textstat.gunning_fog(text)
elif metric == "text_standard":
result = textstat.text_standard(text)
else:
print("ERROR: Please select correct metric!")
result = None
return result
def textstat_stats(text):
doc_length = len(text.split())
flesch_ease = ts.flesch_reading_ease(text) #Flesch Reading Ease Score
flesch_grade = ts.flesch_kincaid_grade(text) #Flesch-Kincaid Grade Level
gfog = ts.gunning_fog(text) # FOG index, also indicates grade level
# smog = ts.smog_index(text) # SMOG index, also indicates grade level, only useful on 30+ sentences
auto_readability = ts.automated_readability_index(text) #approximates the grade level needed to comprehend the text.
cl_index = ts.coleman_liau_index(text) #grade level of the text using the Coleman-Liau Formula.
lw_formula = ts.linsear_write_formula(text) #grade level using the Linsear Write Formula.
dcr_score = ts.dale_chall_readability_score(text) #uses a lookup table of the most commonly used 3000 English words
# text_standard = ts.text_standard(text, float_output=False) # summary of all the grade level functions
syll_count = ts.syllable_count(text, lang='en_US')
syll_count_scaled = syll_count / doc_length
lex_count = ts.lexicon_count(text, removepunct=True)
lex_count_scaled = lex_count / doc_length
idx = ['flesch_ease', 'flesch_grade','gfog',
'auto_readability','cl_index','lw_formula',
'dcr_score',
# 'text_standard',
'syll_count', 'lex_count']
return pd.Series([flesch_ease, flesch_grade, gfog,
auto_readability, cl_index, lw_formula,
dcr_score,
# text_standard,
syll_count_scaled, lex_count_scaled], index = idx)
def get_stats(text):
fre = textstat.flesch_reading_ease(text)
smog = textstat.smog_index(text)
fkg = textstat.flesch_kincaid_grade(text)
cli = textstat.coleman_liau_index(text)
ari = textstat.automated_readability_index(text)
dcr = textstat.dale_chall_readability_score(text)
diff_words = textstat.difficult_words(text)
lwf = textstat.linsear_write_formula(text)
gunn_fog = textstat.gunning_fog(text)
consolidated_score = textstat.text_standard(text)
doc_length = len(text) # think about excluding spaces?
quote_count = text.count('"')
stats = {
"flesch_reading_ease": fre,
"smog_index": smog,
"flesch_kincaid_grade": fkg,
"coleman_liau_index": cli,
"automated_readability_index": ari,
"dale_chall_readability_score": dcr,
"difficult_words": diff_words,
"linsear_write_formula": lwf,
"gunning_fog": gunn_fog,
"consolidated_score": consolidated_score,
"doc_length": doc_length,
"quote_count": quote_count
}
return stats
def score(full):
st.header(textstat.flesch_reading_ease(full))
st.write('Flesch Reading Ease Score')
text = """90-100 Very Easy,70-79 Fairly Easy,60-69 Standard,50-59Fairly Difficult,30-49 Difficult,0-29 Very
Confusing """
st.write(text, key=1)
st.header(textstat.smog_index(full))
st.write('Smog Index Score')
text = "Returns the SMOG index of the given text.This is a grade formula in that a score of 9.3 means that a ninth " \
"grader would be able to read the document.Texts of fewer than 30 sentences are statistically invalid, " \
"because the SMOG formula was normed on 30-sentence samples. textstat requires at least 3 sentences for a " \
"result. "
st.write(text, key=2)
st.header(textstat.dale_chall_readability_score(full))
st.write('Dale Chall Readability Score')
text = """Different from other tests, since it uses a lookup table of the most commonly used 3000 English words.
Thus it returns the grade level using the New Dale-Chall Formula.
4.9 or lower average 4th-grade student or lower
5.0–5.9 average 5th or 6th-grade student
6.0–6.9 average 7th or 8th-grade student
7.0–7.9 average 9th or 10th-grade student
8.0–8.9 average 11th or 12th-grade student
9.0–9.9 average 13th to 15th-grade (college) student"""
st.write(text, key=3)
def terms_and_weights(sample):
sentences = list()
file_path = f"data/Job Bulletins/{sample}"
with open(file_path) as file:
reading_score = textstat.flesch_reading_ease(file_path)
reading_score_2 = textstat.dale_chall_readability_score(file_path)
for line in file:
for l in re.split(r"\.\s|\?\s|\!\s|\n", line):
if l:
sentences.append(l)
cvec = CountVectorizer(stop_words='english',
min_df=3,
max_df=0.5,
ngram_range=(1, 2))
sf = cvec.fit_transform(sentences)
transformer = TfidfTransformer()
transformed_weights = transformer.fit_transform(sf)
weights = np.asarray(transformed_weights.mean(axis=0)).ravel().tolist()
weights_df = pd.DataFrame({
'term': cvec.get_feature_names(),
'weight': weights
})
weights_df = weights_df.sort_values(by='weight', ascending=False).head(10)
myList = {
"term": weights_df.term.tolist(),
"weight": weights_df.weight.tolist(),
"scores": [reading_score, reading_score_2]
}
file.close()
return jsonify(myList)
def calculate_stats(data_folder):
"""Calculate stat of test.json file in a folder"""
data_folder = Path(data_folder)
for dataset in dataset_fields:
print(f"loading {dataset}")
field = dataset_fields[dataset]["text"].strip()
sentences = []
for item in json.load(open(data_folder / dataset / "test.json")):
sentences.append(item[field][-1] if type(item[field]) == list else item[field])
text = " ".join(sentences)
lex_count = textstat.lexicon_count(text)
print(lex_count)
unique_words = count_words(text)
print(f"all unique {len(unique_words)}")
lower_unique_words = count_words(text, casing="lower")
print(f"lowercase unique {len(lower_unique_words)}")
upper_unique_words = count_words(text, casing="upper")
print(f"uppercase unique {len(upper_unique_words)}")
print(f"ratio {len(upper_unique_words) / len(unique_words)}")
text_standard = textstat.text_standard(text, float_output=True)
print(f"text_standard: {text_standard}")
dale_chall_readability_score = textstat.dale_chall_readability_score(text)
print(f"dale_chall_readability_score: {dale_chall_readability_score}")
flesch_kincaid_grade = textstat.flesch_kincaid_grade(text)
print(f"flesch_kincaid_grade: {flesch_kincaid_grade}")
def readability_measures(self, as_dict=False):
"""
Return the BOFIR score as well as other classic readability formulas for the paragraph.
Parameters
----------
as_dict : boolean
Defines if output is a dataframe or dict
Returns
-------
d: DataFrame
DataFrame with the BOFIR score and additional readability measures
"""
flesch = self.flesch
smog = textstat.smog_index(self.paragraph)
dale_chall = textstat.dale_chall_readability_score(self.paragraph)
fog = textstat.gunning_fog(self.paragraph)
bofir_5cat = self.bofir(cat5=True)
bofir_3cat = self.bofir(cat5=False)
d = {
'bofir_5cat': bofir_5cat,
'bofir_3cat': bofir_3cat,
'fog': fog,
'dale_chall': dale_chall,
'smog': smog,
'flesch': flesch
}
if as_dict:
return d
else:
return pd.DataFrame(d, index=['readability_score'])
def seven_test(processed_essay):
"""
score which is assigned to every script in on the basis of some predifened fomulas
These scores are known as readability score.
flesch_score,gunning_index,kincaid_grade,liau_index,automated_readability_index,dale_readability_score,difficult_word,linsear_write
:param processed_essay:
:return:flesch_score,gunning_index,kincaid_grade,liau_index,automated_readability_index,dale_readability_score,difficult_word,linsear_write
"""
flesch_score = ["FS"]
gunning_index = ["GI"]
kincaid_grade = ["KG"]
liau_index = ["LI"]
automated_readability_index = ["ARI"]
dale_readability_score = ["DLS"]
difficult_word = ["DW"]
linsear_write = ["LW"]
for v in processed_essay:
flesch_score.append(textstat.flesch_reading_ease(str(v)))
gunning_index.append(textstat.gunning_fog(str(v)))
kincaid_grade.append(textstat.flesch_kincaid_grade(str(v)))
liau_index.append(textstat.coleman_liau_index(str(v)))
automated_readability_index.append(textstat.automated_readability_index(str(v)))
dale_readability_score.append(textstat.dale_chall_readability_score(str(v)))
difficult_word.append(textstat.difficult_words(str(v)))
linsear_write.append(textstat.linsear_write_formula(str(v)))
return flesch_score,gunning_index,kincaid_grade,liau_index,automated_readability_index,dale_readability_score,difficult_word,linsear_write
def get_readability_score(text, metric="flesch"):
global tknzr, DIFFICULT
text = text.replace("’", "'")
# https://pypi.org/project/textstat/
if metric == "flesch":
return textstat.flesch_reading_ease(text)
elif metric == "smog":
return textstat.smog_index(text)
elif metric == "coleman_liau_index":
return textstat.coleman_liau_index(text)
elif metric == "automated_readability_index":
return textstat.automated_readability_index(text)
elif metric == "dale_chall_readability_score":
return textstat.dale_chall_readability_score(text)
elif metric == "difficult_words":
nb_difficult = 0
nb_easy = 0
for w in set(tknzr.tokenize(text.lower())):
if w not in EASY_WORDS and len(w) >= 6:
nb_difficult += 1
else:
nb_easy += 1
return 100 * nb_difficult / (nb_difficult + nb_easy)
#return textstat.difficult_words(text)#/len(text.split())
elif metric == "linsear_write_formula":
return textstat.linsear_write_formula(text)
elif metric == "gunning_fog":
return textstat.gunning_fog(text)
elif metric == "avg_word_length":
words = tknzr.tokenize(text)
words = [w for w in words if w not in misc_utils.PUNCT]
if len(words) == 0: return 0
return np.average([len(w) for w in words])
def analyze():
print(request)
str_to_read = request.data.decode("utf-8").strip()
report = {
"flesch-reading-ease":
textstat.flesch_reading_ease(str_to_read),
"smog-index":
textstat.smog_index(str_to_read),
"flesch-kincaid-grade":
textstat.flesch_kincaid_grade(str_to_read),
"coleman-liau-index":
textstat.coleman_liau_index(str_to_read),
"automated-readability-index":
textstat.automated_readability_index(str_to_read),
"dale-chall-readability-score":
textstat.dale_chall_readability_score(str_to_read),
"difficult-words":
textstat.difficult_words(str_to_read),
"linsear-write-formula":
textstat.linsear_write_formula(str_to_read),
"gunning-fog":
textstat.gunning_fog(str_to_read),
"text-standard":
textstat.text_standard(str_to_read)
}
return decorate_response(jsonify(report))
def readability(queries):
scores = pd.DataFrame(columns=[
'Flesch', 'Smog', 'Flesch grade', 'Coleman', 'Automated', 'Dale',
'Difficult', 'Linsear', 'Gunning', 'Text Standard'
])
scores = {
'Flesch': [],
'Smog': [],
'Flesch grade': [],
'Coleman': [],
'Automated': [],
'Dale': [],
'Difficult': [],
'Linsear': [],
'Gunning': [],
'Text Standard': []
}
for line in queries:
# results = readability.getmeasures(line, lang='en')
# frescores.append(results['readability grades']['FleschReadingEase'])
# line = 'yao family wines . yao family wines is a napa valley producer founded in 2011 by yao ming , the chinese-born , five-time nba all star . now retired from the houston rockets , yao ming is the majority owner in yao family wines , which has entered the wine market with a luxury cabernet sauvignon sourced from napa valley vineyards .'
scores['Flesch'].append(textstat.flesch_reading_ease(line))
scores['Smog'].append(textstat.smog_index(line))
scores['Flesch grade'].append(textstat.flesch_kincaid_grade(line))
scores['Coleman'].append(textstat.coleman_liau_index(line))
scores['Automated'].append(textstat.automated_readability_index(line))
scores['Dale'].append(textstat.dale_chall_readability_score(line))
scores['Difficult'].append(textstat.difficult_words(line))
scores['Linsear'].append(textstat.linsear_write_formula(line))
scores['Gunning'].append(textstat.gunning_fog(line))
scores['Text Standard'].append(
textstat.text_standard(line, float_output=True))
return scores
def getReadabilityMetrics(test_data):
'''
for a given article IN TEXT FORMAT, returns its readability metrics
Uses textstat library, please install it
'''
metric = {
"flesch_reading_ease":
textstat.flesch_reading_ease(test_data),
"smog_index":
textstat.smog_index(test_data),
"flesch_kincaid_grade":
textstat.flesch_kincaid_grade(test_data),
"coleman_liau_index":
textstat.coleman_liau_index(test_data),
"automated_readability_index":
textstat.automated_readability_index(test_data),
"dale_chall_readability_score":
textstat.dale_chall_readability_score(test_data),
"difficult_words":
textstat.difficult_words(test_data),
"linsear_write_formula":
textstat.linsear_write_formula(test_data),
"gunning_fog":
textstat.gunning_fog(test_data),
"text_standard":
textstat.text_standard(test_data)
}
return metric
def readability_scores_mp(data):
result_dict, idx, text = data
# flesch_reading_ease = textstat.flesch_reading_ease(text)
flesch_kincaid_grade = textstat.flesch_kincaid_grade(text)
dale_chall_readability_score = textstat.dale_chall_readability_score(text)
result_dict[idx] = [flesch_kincaid_grade, dale_chall_readability_score]
def getReadability(df):
import textstat
df['ARI'] = df.headline_text.apply(
lambda x: textstat.automated_readability_index(x))
df['DCR'] = df.headline_text.apply(
lambda x: textstat.dale_chall_readability_score(x))
df['TS'] = df.headline_text.apply(
lambda x: textstat.text_standard(x, float_output=True))
return df
def metrics(sentence):
fk = round(flesch_kincaid_grade(sentence), 3)
gf = round(gunning_fog(sentence), 3)
dc = round(dale_chall_readability_score(sentence), 3)
fk_label = grade_label(round(fk))
gf_label = grade_label(round(gf))
dc_label = grade_label(dale_chall_norm(round(dc)))
return (fk, gf, dc, fk_label, gf_label, dc_label)
def compute_readability_stats(text):
"""
Compute reading statistics of the given text
Reference: https://github.com/shivam5992/textstat
Parameters
==========
text: str, input section or abstract text
"""
try:
readability_dict = {
'flesch_reading_ease':
textstat.flesch_reading_ease(text),
'smog':
textstat.smog_index(text),
'flesch_kincaid_grade':
textstat.flesch_kincaid_grade(text),
'coleman_liau_index':
textstat.coleman_liau_index(text),
'automated_readability_index':
textstat.automated_readability_index(text),
'dale_chall':
textstat.dale_chall_readability_score(text),
'difficult_words':
textstat.difficult_words(text),
'linsear_write':
textstat.linsear_write_formula(text),
'gunning_fog':
textstat.gunning_fog(text),
'text_standard':
textstat.text_standard(text),
'n_syllable':
textstat.syllable_count(text),
'avg_letter_per_word':
textstat.avg_letter_per_word(text),
'avg_sentence_length':
textstat.avg_sentence_length(text)
}
except:
readability_dict = {
'flesch_reading_ease': None,
'smog': None,
'flesch_kincaid_grade': None,
'coleman_liau_index': None,
'automated_readability_index': None,
'dale_chall': None,
'difficult_words': None,
'linsear_write': None,
'gunning_fog': None,
'text_standard': None,
'n_syllable': None,
'avg_letter_per_word': None,
'avg_sentence_length': None
}
return readability_dict
def score(self, strText):
self.automated_readability_index = textstat.automated_readability_index(
strText)
self.str_automated_readability_index = self.grade(
self.automated_readability_index)
self.coleman_liau_index = textstat.coleman_liau_index(strText)
self.str_coleman_liau_index = self.grade(self.coleman_liau_index)
self.dale_chall_readability_score = textstat.dale_chall_readability_score(
strText)
if self.dale_chall_readability_score >= 9.0:
self.str_dale_chall_readability_score = ' | ' + '13th to 15th grade (college)'
elif self.dale_chall_readability_score >= 8.0:
self.str_dale_chall_readability_score = ' | ' + '11th to 12th grade'
elif self.dale_chall_readability_score >= 7.0:
self.str_dale_chall_readability_score = ' | ' + '9th to 10th grade'
elif self.dale_chall_readability_score >= 6.0:
self.str_dale_chall_readability_score = ' | ' + '7th to 8th grade'
elif self.dale_chall_readability_score >= 5.0:
self.str_dale_chall_readability_score = ' | ' + '5th to 6th grade'
else:
self.str_dale_chall_readability_score = ' | ' + '4th grade or lower'
self.difficult_words = textstat.difficult_words(strText)
self.flesch_kincaid_grade = textstat.flesch_kincaid_grade(strText)
self.str_flesch_kincaid_grade = self.grade(self.flesch_kincaid_grade)
self.flesch_reading_ease = textstat.flesch_reading_ease(strText)
if self.flesch_reading_ease >= 90:
self.str_flesch_reading_ease = ' | ' + 'Very Easy'
elif self.flesch_reading_ease >= 80:
self.str_flesch_reading_ease = ' | ' + 'Easy'
elif self.flesch_reading_ease >= 70:
self.str_flesch_reading_ease = ' | ' + 'Fairly Easy'
elif self.flesch_reading_ease >= 60:
self.str_flesch_reading_ease = ' | ' + 'Standard'
elif self.flesch_reading_ease >= 50:
self.str_flesch_reading_ease = ' | ' + 'Fairly Difficult'
elif self.flesch_reading_ease >= 30:
self.str_flesch_reading_ease = ' | ' + 'Difficult'
else:
self.str_flesch_reading_ease = ' | ' + 'Very Confusing'
self.gunning_fog = textstat.gunning_fog(strText)
self.str_gunning_fog = self.grade(self.gunning_fog)
self.linsear_write_formula = textstat.linsear_write_formula(strText)
self.str_linsear_write_formula = self.grade(self.linsear_write_formula)
self.smog_index = textstat.smog_index(strText)
self.str_smog_index = self.grade(self.smog_index)
self.text_standard = textstat.text_standard(strText)
def generate_score(self, text):
self.flesch_reading_grade = ts.flesch_reading_ease(text)
self.flesch_reading_grade_consensus = readability_test_consensus(self.flesch_reading_grade, flesch_ease_grading_system)
self.flesch_kincaid_grade = ts.flesch_kincaid_grade(text)
self.flesch_kincaid_grade_consensus = readability_test_consensus(self.flesch_kincaid_grade, us_grade_level_system_age)
self.dale_chall_grade = ts.dale_chall_readability_score(text)
self.dale_chall_grade_consensus = readability_test_consensus(self.dale_chall_grade, dale_chall_system)
self.smog_grade = ts.smog_index(text)
self.ari_grade = ts.automated_readability_index(text)
""" self.ari_grade_consensus = readability_test_consensus(self.ari_grade, us_grade_level_system_level) """
self.coleman_liau_grade = ts.coleman_liau_index(text)
pass
def process(self, df):
t0 = time()
print("\n---Generating Readability Features:---\n")
def lexical_diversity(text):
words = nltk.tokenize.word_tokenize(text.lower())
word_count = len(words)
vocab_size = len(set(words))
diversity_score = vocab_size / word_count
return diversity_score
def get_counts(text, word_list):
words = nltk.tokenize.word_tokenize(text.lower())
count = 0
for word in words:
if word in word_list:
count += 1
return count
df['flesch_reading_ease'] = df['articleBody'].map(lambda x: textstat.flesch_reading_ease(x))
df['smog_index'] = df['articleBody'].map(lambda x: textstat.smog_index(x))
df['flesch_kincaid_grade'] = df['articleBody'].map(lambda x: textstat.flesch_kincaid_grade(x))
df['coleman_liau_index'] = df['articleBody'].map(lambda x: textstat.coleman_liau_index(x))
df['automated_readability_index'] = df['articleBody'].map(lambda x: textstat.automated_readability_index(x))
df['dale_chall_readability_score'] = df['articleBody'].map(lambda x: textstat.dale_chall_readability_score(x))
df['difficult_words'] = df['articleBody'].map(lambda x: textstat.difficult_words(x))
df['linsear_write_formula'] = df['articleBody'].map(lambda x: textstat.linsear_write_formula(x))
df['gunning_fog'] = df['articleBody'].map(lambda x: textstat.gunning_fog(x))
df['i_me_myself'] = df['articleBody'].apply(get_counts,args = (['i', 'me', 'myself'],))
df['punct'] = df['articleBody'].apply(get_counts,args = ([',','.', '!', '?'],))
df['lexical_diversity'] = df['articleBody'].apply(lexical_diversity)
feats = ['flesch_reading_ease', 'smog_index', 'flesch_kincaid_grade',
'coleman_liau_index', 'automated_readability_index',
'dale_chall_readability_score', 'difficult_words', 'linsear_write_formula',
'gunning_fog', 'i_me_myself', 'punct', 'lexical_diversity'
]
outfilename_xReadable = df[feats].values
with open('../saved_data/read.pkl', 'wb') as outfile:
pickle.dump(feats, outfile, -1)
pickle.dump(outfilename_xReadable, outfile, -1)
print ('readable features saved in read.pkl')
print('\n---Readability Features is complete---')
print("Time taken {} seconds\n".format(time() - t0))
return 1
def cal_readability(target, source):
import pandas as pd
tf_r_es = [textstat.flesch_reading_ease(t) for t in target]
tf_k_gs = [textstat.flesch_kincaid_grade(t) for t in target]
td_c_rs = [textstat.dale_chall_readability_score(t) for t in target]
sf_r_es = [textstat.flesch_reading_ease(t) for t in source]
sf_k_gs = [textstat.flesch_kincaid_grade(t) for t in source]
sd_c_rs = [textstat.dale_chall_readability_score(t) for t in source]
diff_r_es = [np.abs(tf_r_es[i] - sf_r_es[i]) for i in range(len(tf_r_es))]
diff_k_gs = [np.abs(tf_k_gs[i] - sf_k_gs[i]) for i in range(len(tf_k_gs))]
difd_c_rs = [np.abs(td_c_rs[i] - sd_c_rs[i]) for i in range(len(td_c_rs))]
return {"Flesch ease mean gen": np.mean(tf_r_es), \
"Flesch ease mean orig": np.mean(sf_r_es), \
"Flesch ease mean diff": np.mean(diff_r_es), \
"Flesch grade mean gen": np.mean(tf_k_gs), \
"Flesch grade mean orig": np.mean(sf_k_gs), \
"Flesch grade mean diff": np.mean(diff_k_gs), \
"Dale Chall Readability V2 mean gen": np.mean(td_c_rs), \
"Dale Chall Readability V2 mean orig": np.mean(sd_c_rs), \
"Dale Chall Readability V2 mean diff": np.mean(difd_c_rs), \
},\
\
{"Flesch ease std dev gen": np.std(tf_r_es), \
"Flesch ease std dev orig": np.std(sf_r_es), \
"Flesch ease std dev diff": np.std(diff_r_es), \
"Flesch grade std dev gen": np.std(tf_k_gs), \
"Flesch grade std dev orig": np.std(sf_k_gs), \
"Flesch grade std dev diff": np.std(diff_k_gs), \
"Dale Chall Readability V2 std dev gen": np.std(td_c_rs),\
"Dale Chall Readability V2 std dev orig": np.std(sd_c_rs),\
"Dale Chall Readability V2 std dev diff": np.std(difd_c_rs)\
}
def readability_scores(self, text):
self.ari = textstat.automated_readability_index(text)
self.flesch_kincaid_grade = textstat.flesch_kincaid_grade(text)
self.coleman_liau_index = textstat.coleman_liau_index(text)
self.dale_chall_readability_score = textstat.dale_chall_readability_score(
text)
self.flesch_reading_ease = textstat.flesch_reading_ease(text)
self.gunning_fog = textstat.gunning_fog(text)
self.linsear_write_formula = textstat.linsear_write_formula(text)
self.lix = textstat.lix(text)
self.rix = textstat.rix(text)
self.smog_index = textstat.smog_index(text)
self.text_standard = textstat.text_standard(text)
def get_readability_stats(text):
return {
'flesch_reading_ease': textstat.flesch_reading_ease(text),
'smog_index': textstat.smog_index(text),
'flesch_kincaid_grade': textstat.flesch_kincaid_grade(text),
'coleman_liau_index': textstat.coleman_liau_index(text),
'automated_readability_index':
textstat.automated_readability_index(text),
'dale_chall_readability_score':
textstat.dale_chall_readability_score(text),
'linsear_write_formula': textstat.linsear_write_formula(text),
'gunning_fog': textstat.gunning_fog(text),
'text_standard': textstat.text_standard(text, float_output=True),
}
def lisibilty(text):
f_lis = ([
textstat.syllable_count(str(text), lang='en_arabic'),
textstat.lexicon_count(str(text), removepunct=True),
textstat.sentence_count(str(text)),
textstat.flesch_reading_ease(str(text)),
textstat.flesch_kincaid_grade(str(text)),
textstat.gunning_fog(str(text)),
textstat.smog_index(str(text)),
textstat.automated_readability_index(str(text)),
textstat.coleman_liau_index(str(text)),
textstat.linsear_write_formula(str(text)),
textstat.dale_chall_readability_score(str(text))
])
return f_lis
def analyze_vocab(text):
return {
'num_words': textstat.lexicon_count(text),
'flesch_reading_ease': textstat.flesch_reading_ease(text),
'smog_index': textstat.smog_index(text),
'flesch_kincaid_grade': textstat.flesch_kincaid_grade(text),
'coleman_liau_index': textstat.coleman_liau_index(text),
'automated_readability_index':
textstat.automated_readability_index(text),
'dale_chall_readability_score':
textstat.dale_chall_readability_score(text),
'difficult_words': textstat.difficult_words(text),
'linsear_write_formula': textstat.linsear_write_formula(text),
'gunning_fog': textstat.gunning_fog(text),
'text_standard': textstat.text_standard(text, float_output=True)
}
def vocab_check(text):
#Construct dictionary
vocab_results = {'dale_chall_readability_score': dale_chall_readability_score(text),
'smog_index': smog_index(text), 'gunning_fog': gunning_fog(text),
'flesch_reading_ease': flesch_reading_ease(text),
'flesch_kincaid_grade': flesch_kincaid_grade(text),
'linsear_write_formula': linsear_write_formula(text),
'coleman_liau_index': coleman_liau_index(text),
'automated_readability_index': automated_readability_index(text),
'yule_vocab_richness': yule(text),
'total_score': text_standard(text, float_output=True)}
diff_words, easy_word_dict = difficult_words(text)
return(vocab_results, diff_words, easy_word_dict)
def add_features(row):
'''Feature engineering via NLP.'''
text = row.text
doc = nlp(text)
lemmas = list()
entities = list()
for token in doc:
if token.text == ':':
row['has_colon'] = 1
if token.text == ';':
row['has_semicolon'] = 1
if token.text == '-':
row['has_dash'] = 1
if token.text.lower() == 'whom':
row['whom'] = 1
if token.text[-3:] == 'ing':
row['num_ings'] += 1
if token.text.lower() == 'had':
row['has_had'] = 1
pos = token.pos_
row[pos] += 1
if token.is_stop or not token.is_alpha:
continue
lemma = token.lemma_.strip().lower()
if lemma:
lemmas.append(lemma)
for ent in doc.ents:
entities.append(ent.text)
lemmas = ' '.join(lemmas)
blob = TextBlob(text)
row['subjectivity'] = blob.sentiment.subjectivity
row['polarity'] = blob.sentiment.polarity
row['starts_conj'] = int(doc[0].pos_ == 'CONJ')
row['ends_prep'] = int(doc[0].pos_ == 'PREP')
row['entities'] = entities
row['lemmas'] = lemmas
row['raw_text_length'] = len(text)
row['num_words'] = len(doc)
row['avg_word_len'] = row.raw_text_length / row.num_words
row['vector_avg'] = np.mean(nlp(lemmas).vector)
row['num_ings'] /= row['num_words']
row['rhyme_frequency'] = rhyme_frequency(row['text'])
row['dale_chall'] = textstat.dale_chall_readability_score(row['text'])
row['FleischReadingEase'] = textstat.flesch_reading_ease(row['text'])
row['lexicon'] = textstat.lexicon_count(row['text'])
row['word_diversity'] = row.lexicon / row.num_words
return row
def textstat_stats(text):
difficulty = textstat.flesch_reading_ease(text)
grade_difficulty = textstat.flesch_kincaid_grade(text)
gfog = textstat.gunning_fog(text)
smog = textstat.smog_index(text)
ari = textstat.automated_readability_index(text)
cli = textstat.coleman_liau_index(text)
lwf = textstat.linsear_write_formula(text)
dcrs = textstat.dale_chall_readability_score(text)
idx = [
'difficulty', 'grade_difficulty', 'gfog', 'smog', 'ari', 'cli', 'lwf',
'dcrs'
]
return pd.Series(
[difficulty, grade_difficulty, gfog, smog, ari, cli, lwf, dcrs],
index=idx)
def get_readability_features(self):
sent_tokens = text_tokenizer(self.raw_text,
replace_url_flag=True,
tokenize_sent_flag=True)
sentences = [' '.join(sent) + '\n' for sent in sent_tokens]
sentences = ''.join(sentences)
self.syllable_count = textstat.syllable_count(sentences)
self.flesch_reading_ease = textstat.flesch_reading_ease(sentences)
self.flesch_kincaid_grade = textstat.flesch_kincaid_grade(sentences)
self.fog_scale = textstat.gunning_fog(sentences)
self.smog = textstat.smog_index(sentences)
self.automated_readability = textstat.automated_readability_index(
sentences)
self.coleman_liau = textstat.coleman_liau_index(sentences)
self.linsear_write = textstat.linsear_write_formula(sentences)
self.dale_chall_readability = textstat.dale_chall_readability_score(
sentences)
self.text_standard = textstat.text_standard(sentences)
def score_text(self, test_data):
score = {}
score['flesch_reading_ease'] = textstat.flesch_reading_ease(test_data)
score['smog_index'] = textstat.smog_index(test_data)
score['flesch_kincaid_grade'] = textstat.flesch_kincaid_grade(
test_data)
score['coleman_liau_index'] = textstat.coleman_liau_index(test_data)
score[
'automated_readability_index'] = textstat.automated_readability_index(
test_data)
score[
'dale_chall_readability_score'] = textstat.dale_chall_readability_score(
test_data)
score['difficult_words'] = textstat.difficult_words(test_data)
score['linsear_write_formula'] = textstat.linsear_write_formula(
test_data)
score['gunning_fog'] = textstat.gunning_fog(test_data)
score['text_standard'] = textstat.text_standard(test_data)
return score
def test_dale_chall_readability_score():
score = textstat.dale_chall_readability_score(long_test)
assert score == 6.87
