def ngram(wordex, period, ddk=None):
if " " in word:
bigram = word.split()[:2]
res = nb.bigram(first=bigram[0],
second=bigram[1],
ddk=ddk,
period=period)
else:
res = nb.unigram(word, period=period, ddk=ddk)
return res
def sumword(words, period, media='bok'):
wordlist = [x.strip() for x in words.split(',')]
# check if trailing comma, or comma in succession, if so count comma in
if '' in wordlist:
wordlist = [','] + [y for y in wordlist if y != '']
ref = pd.concat(
[nb.unigram(w, media=media, period=period) for w in wordlist],
axis=1).sum(axis=1)
ref.columns = ["tot"]
return ref
def ngram(word, ddk, subject, period):
if " " in word:
bigram = word.split()[:2]
res = nb.bigram(first=bigram[0],
second=bigram[1],
ddk=ddk,
topic=subject,
period=period)
else:
res = nb.unigram(word, ddk=ddk, topic=subject, period=period)
return res
def sumword(words, period, media='bok', lang='nob'):
wordlist = [x.strip() for x in words.split(',')]
# check if trailing comma, or comma in succession, if so count comma in
if '' in wordlist:
wordlist = [','] + [y for y in wordlist if y != '']
ref = pd.concat([
nb.unigram(w, media=media, period=period, lang=lang) for w in wordlist
],
axis=1).sum(axis=1)
ref.index = pd.to_datetime(ref.index, format='%Y')
return ref
def ngavis(word, period):
try:
if " " in word:
bigram = word.split()[:2]
res = nb.frame(
nb.bigram(first=bigram[0],
second=bigram[1],
period=period,
media='avis'), word)
else:
res = nb.frame(nb.unigram(word, period=period, media='avis'), word)
#st.write(res.head())
except:
res = pd.DataFrame()
return res
def ngbok(word, period, ddk=None, lang='nob'):
try:
if " " in word:
bigram = word.split()[:2]
res = nb.frame(
nb.bigram(first=bigram[0],
second=bigram[1],
ddk=ddk,
period=period,
media='bok',
lang=lang), word)
else:
res = nb.frame(
nb.unigram(word,
period=period,
ddk=ddk,
media='bok',
lang=lang), word)
except:
res = pd.DataFrame()
return res
def ngavis(x, period):
try:
r = nb.frame(nb.unigram(x, period, media='avis'), x)
except:
r = pd.DataFrame()
return r
def ngbok(x, period, ddk=None):
try:
r = nb.frame(nb.unigram(x, period, media='bok', ddk=ddk), x)
except:
r = pd.DataFrame()
return r