def plot(game_name, game_id):
dict = {}
comments = DbUtil.getAllResult("select * from comment where game_id = %s" % game_id)
for comment in comments:
result = jieba.analyse.extract_tags(comment[2], topK=3)
for word in result:
if len(word) < 2:
continue
elif word in stop:
continue
if word not in dict:
dict[word] = 1
else:
dict[word] += 1
print(dict)
swd = sorted(dict.items(), key=itemgetter(1), reverse=True)
swd = swd[1:50]
tags = make_tags(swd,
minsize=30,
maxsize=120,
colors=random.choice(list(COLOR_SCHEMES.values())))
create_tag_image(tags,
'c:/%s.png' % game_name,
background=(0, 0, 0, 255),
size=(900, 600),
fontname='SimHei')
print('having save file to dick')
def plot(game_name, game_id):
dict = {}
comments = DbUtil.getAllResult("select * from comment where game_id = %s" %
game_id)
for comment in comments:
result = jieba.analyse.extract_tags(comment[2], topK=3)
for word in result:
if len(word) < 2:
continue
elif word in stop:
continue
if word not in dict:
dict[word] = 1
else:
dict[word] += 1
print(dict)
swd = sorted(dict.items(), key=itemgetter(1), reverse=True)
swd = swd[1:50]
tags = make_tags(swd,
minsize=30,
maxsize=120,
colors=random.choice(list(COLOR_SCHEMES.values())))
create_tag_image(tags,
'c:/%s.png' % game_name,
background=(0, 0, 0, 255),
size=(900, 600),
fontname='SimHei')
print('having save file to dick')
def cut():
comments = DbUtil.getAllResult("select * from comment limit 300000")
file = open("test1.txt", "w",encoding="utf-8")
for comment in comments:
list = []
result = jieba.cut(comment[2])
for word in result:
if word not in stop and word != ' ':
list.append(word)
if list:
file.write(" ".join(list))
file.write("\n")
file.close()
pass
def plot(game_name, game_id):
dict = {}
comments = DbUtil.getAllResult("select * from comment where game_id = %s" %
game_id)
for comment in comments:
result = jieba.analyse.extract_tags(comment[2], topK=3)
for word in result:
if len(word) < 2:
continue
elif word in stop:
continue
if word not in dict:
dict[word] = 1
else:
dict[word] += 1
swd = sorted(dict.items(), key=itemgetter(1), reverse=True)
swd = swd[1:50]
tags = make_tags(swd,
minsize=30,
maxsize=100,
colors=random.choice(list(COLOR_SCHEMES.values())))
create_tag_image(tags,
'C:/Users/Administrator/Desktop/%s_%s.png' %
(game_name, game_id),
background=(0, 0, 0, 255),
size=(900, 600),
fontname='SimHei')
print('create file ---%s' % game_name)
# dict = {}
#
# for (k, v) in swd:
# dict[k] = v
# print('INSERT INTO keyword (game_id, keyword) VALUES (%s, "%s"' % (game_id, str(dict)))
# cur.execute('INSERT INTO keyword (game_id, keyword) VALUES (%s, "%s")' % (game_id, str(dict)))
# conn.commit()
word = DbUtil.getOneResult('select keyword from keyword limit 1')
print(eval(word[0]))
def plot(game_name, game_id):
dict = {}
comments = DbUtil.getAllResult("select * from comment where game_id = %s" % game_id)
for comment in comments:
result = jieba.analyse.extract_tags(comment[2], topK=3)
for word in result:
if len(word) < 2:
continue
elif word in stop:
continue
if word not in dict:
dict[word] = 1
else:
dict[word] += 1
swd = sorted(dict.items(), key=itemgetter(1), reverse=True)
swd = swd[1:50]
tags = make_tags(swd,
minsize=30,
maxsize=100,
colors=random.choice(list(COLOR_SCHEMES.values())))
create_tag_image(tags,
'C:/Users/Administrator/Desktop/%s_%s.png' % (game_name, game_id),
background=(0, 0, 0, 255),
size=(900, 600),
fontname='SimHei')
print('create file ---%s' % game_name)
# dict = {}
#
# for (k, v) in swd:
# dict[k] = v
# print('INSERT INTO keyword (game_id, keyword) VALUES (%s, "%s"' % (game_id, str(dict)))
# cur.execute('INSERT INTO keyword (game_id, keyword) VALUES (%s, "%s")' % (game_id, str(dict)))
# conn.commit()
word = DbUtil.getOneResult('select keyword from keyword limit 1')
print(eval(word[0]))
def toDict(list):
return dict([(word, True) for word in list if word in best_words])
def features(feature_extraction_method):
Features = []
for i in lists:
words = feature_extraction_method(i) # 为积极文本赋予"pos"
Features.append(words)
return Features
if __name__ == '__main__':
getStop()
comments = list(DbUtil.getAllResult("select * from comment"))
shuffle(comments)
conn, cur = DbUtil.getConn()
for i in range(0, 1065000, 1000):
print(i)
for comment in comments[i:i + 1000]:
print('UPDATE comment set comment_time = "%s" where id =%s' %
(Ran.getTime(int(i / 1000 + 1) % 30 + 1), comment[0]))
cur.execute('UPDATE comment set comment_time = "%s" where id =%s' %
(Ran.getTime(int(i / 1000 + 1) % 30 + 1), comment[0]))
conn.commit()
DbUtil.close(conn, cur)
# lists = []
# for comment in comments:
# list = []
create_tag_image(tags,
'c:/%s.png' % game_name,
background=(0, 0, 0, 255),
size=(900, 600),
fontname='SimHei')
print('having save file to dick')
if __name__ == '__main__':
f = open("../StopWords.txt", encoding="utf-8")
jieba.load_userdict("c:/dict.txt")
while True:
line = f.readline().replace("\n", '')
if not line:
break
stop.append(line)
games = DbUtil.getAllResult(
"select game_id,games.game_name from `comment` join games on game_id = games.id GROUP BY game_id ORDER BY count(game_id) desc limit 50"
)
l = []
for game in games:
if game[1] not in l:
plot(game[1], game[0])
l.append(game[1])
colors=random.choice(list(COLOR_SCHEMES.values())))
create_tag_image(tags,
'c:/%s.png' % game_name,
background=(0, 0, 0, 255),
size=(900, 600),
fontname='SimHei')
print('having save file to dick')
if __name__ == '__main__':
f = open("../StopWords.txt", encoding="utf-8")
jieba.load_userdict("c:/dict.txt")
while True:
line = f.readline().replace("\n", '')
if not line:
break
stop.append(line)
games = DbUtil.getAllResult("select game_id,games.game_name from `comment` join games on game_id = games.id GROUP BY game_id ORDER BY count(game_id) desc limit 50")
l =[]
for game in games:
if game[1] not in l:
plot(game[1], game[0])
l.append(game[1])
if __name__ == '__main__':
getStop()
pos = pickle.load(open("pos_review.pkl", 'rb'))
neg = pickle.load(open("neg_review.pkl", 'rb'))
print(len(pos))
print(len(neg))
while [] in pos:
pos.pop(pos.index([]))
while [] in neg:
neg.pop(neg.index([]))
comments = list(
DbUtil.getAllResult("select * from comment limit 10000 offset 600000"))
shuffle(comments)
comments = comments[:100]
lists = []
for comment in comments:
list = []
result = jieba.cut(comment[2])
for word in result:
if word not in stop and word != ' ':
list.append(word)
lists.append(list)
count = 0
frame = tkinter.Tk()
print(neg)
if __name__ == '__main__':
getStop()
pos = pickle.load(open("pos_review.pkl", 'rb'))
neg = pickle.load(open("neg_review.pkl", 'rb'))
print(len(pos))
print(len(neg))
while [] in pos:
pos.pop(pos.index([]))
while [] in neg:
neg.pop(neg.index([]))
comments = list(DbUtil.getAllResult("select * from comment limit 10000 offset 600000"))
shuffle(comments)
comments = comments[:100]
lists = []
for comment in comments:
list = []
result = jieba.cut(comment[2])
for word in result:
if word not in stop and word != ' ':
list.append(word)
lists.append(list)
count = 0
frame = tkinter.Tk()
def toDict(list):
return dict([(word, True) for word in list if word in best_words])
def features(feature_extraction_method):
Features = []
for i in lists:
words = feature_extraction_method(i) # 为积极文本赋予"pos"
Features.append(words)
return Features
if __name__ == '__main__':
getStop()
comments = list(DbUtil.getAllResult("select * from comment"))
shuffle(comments)
conn, cur = DbUtil.getConn()
for i in range(0,1065000,1000):
print(i)
for comment in comments[i:i+1000]:
print('UPDATE comment set comment_time = "%s" where id =%s' %(Ran.getTime(int(i/1000+1)%30+1),comment[0]))
cur.execute('UPDATE comment set comment_time = "%s" where id =%s' %(Ran.getTime(int(i/1000+1)%30+1),comment[0]))
conn.commit()
DbUtil.close(conn,cur)
# lists = []
# for comment in comments:
# list = []
# result = jieba.cut(comment[2])
# for word in result: