編集履歴

質問編集履歴

著作権で問題がでないように、必要箇所のみ掲載

2018/10/07 13:18

投稿

giro1975

スコア37

test CHANGED Viewed

File without changes

test CHANGED Viewed

@@ -28,310 +28,6 @@
-```ここに言語を入力
-#twitkey呼び出し
-import twitkey #自作の認証キー
-import tweepy
-CK = twitkey.twkey['cons_key']
-CS = twitkey.twkey['cons_sec']
-AK = twitkey.twkey['accto_key']
-AT = twitkey.twkey['accto_sec']
-```
-```ここに言語を入力
-#twitter データ収集用プログラム
-def get_twitter_api(CK,CS,AK,AT):
-    auth = tweepy.OAuthHandler(CK,CS) #OAuthHandler
-    auth.set_access_token(AK,AT)
-    api = tweepy.API(auth)
-    return api
-def get_tweet(user_id,count):
-    API = get_twitter_api(CK,CS,AK,AT)
-    data = API.user_timeline(id=user_id,count=count)
-    tweets = []
-    for tweet in data:
-        tweets.append(format_text(tweet.text))
-        return tweets,data[-1].id
-def get_tweet_with_id(user_id,count,next_max_id):
-    API = get_twitter_api(CK,CS,AK,AT)
-    data = API.user_timeline(id=user_id,count=count,max_id=next_max_id-1)
-    tweets = []
-    for tweet in data:
-        tweets.append(format_text(tweet.text))
-        if len(data)  == 0:
-            return tweets,0
-        return tweets,data[-1].id
-```
-```ここに言語を入力
-# tweet 収集用プログラム
-def get_positive_tweets():
-    pos_tweets = []
-    positive_ids =[
-        "positive_bot_00",
-        "positivekk_bot",
-        "botpositive",
-        "positive_mot",
-        "kami_positive",
-        "positive_bot",
-        "jinseiplusbot",
-        "syuzou_genki",
-        "genki_kotoba_m"
-        ]
-    for pos_id in positive_ids:
-        tmp,max_id = get_tweet(pos_id,200)
-        tmp = list(set(tmp))
-        for i in range(len(tmp)):
-            pos_tweets.append(tmp[i])
-    while True:
-        tmp,max_id = get_tweet_with_id(pos_id,100,max_id)
-        tmp = list(set(tmp))
-        for i in range(len(tmp)):
-            pos_tweets.append(tmp[i])
-        if max_id == 0:
-            break
-        return pos_tweets
-def get_negative_tweets():
-    neg_tweets = []
-    negative_ids = [
-        "negatizibu_bot",
-        "immydream19",
-        "lewyDanf",
-        "positive_act_me",
-        "pgmtmw",
-        "yamiki_bot",
-        "cool_aroma",
-        "nega_bot",
-        "negativebot",
-        "H4Za5",
-        "ymibot"
-        ]
-    for neg_id in negative_ids:
-        tmp,max_id = get_tweet(neg_id,200)
-        tmp = list(set(tmp))
-        for i in range(len(tmp)):
-            neg_tweets.append(tmp[i])
-    while True:
-        tmp,max_id = get_tweet_with_id(neg_id,100,max_id)
-        tmp = list(set(tmp))
-        for i in range(len(tmp)):
-            neg_tweets.append(tmp[i])
-        if max_id == 0:
-            break
-        return neg_tweets
-```
-```ここに言語を入力
-#url,必要のない文字列を削除するプログラム
-import re
-def format_text(text):
-    text = re.sub(r'https?://[\w/:%#$&?(/)~.=+\-...]+',"",'text')
-    text = re.sub('RT',"",text)
-    text = re.sub('お気に入り',"",text)
-    text = re.sub('まとめ',"",text)
-    text = re.sub(r'[!-~]','""',text) #半角記号,数字,英字
-    text = re.sub(r'[:-@]',"",text) #全角記号
-    text = re.sub('\n',"",text) #改行文字
-    return text
-```
-```ここに言語を入力
-#形態素解析
-def tokenize(tweets):
-    t = Tokenizer()
-    tokennaized_tweets = []
-    for tw in tweets:
-        tokens = t.tokenize(tw)
-        tmp = ""
-        for token in tokens:
-            noun_flag = 0
-            partOfSpeech = token.part_of_speech.split(",")[0]
-            if partOfSpeech == "名刺":
-                noun_flag = 1
-            if partOfSpeech == "動詞":
-                noun_flag = 1
-            if partOfSpeech == "形容詞":
-                noun_flag = 1
-            if partOfSpeech == "形容動詞":
-                noun_flag = 1
-            if partOfSpeech == "感動詞":
-                noun_flag = 1
-            if noun_flag == 1:
-                tmp += token.surface + ""
-        tmp = tmp.rfstrip(" ")
-        tokenized_tweets.append(tmp)
-    return tokenized_tweets
-```
 ```ここに言語を入力
 #ポジティブ、ネガティブなツイートを取得、そして整形 ※ここでエラー