コード例 #1
0
def countwords(results):
   count=dict()
   for value in results["results"]:
    #get the text from each tweet
    text=value["text"]
    #correct ASCII issues
    text = filter(lambda x: x in string.printable, text)
    #add space at the end of the line
    text = text + " "
    #stem text
    text1=porterstemmer.stemWords(text)
    #find all strings that only contains alphabets from the stemming text 
    words=re.findall("[A-Za-z]+",text1)
    for word in swords:
        t=words.count(word)
        if word in count:
            count[word]+=t
        else:
            count[word]=t
   return count
コード例 #2
0
list=open("sentimentList.txt")

#create a dictionary to store stemming words and values
swords=dict()

for line in list:
    line=line.strip()
    #split the line into words by the last comma
    words=line.rsplit(",")
    #words[0] is the sentiment word or expression
    word=words[0]  
    
    #check if it is an actual word
    if not word.isalpha(): continue
    #if it is a word, then run porterstemmer to stem the words
    line1=porterstemmer.stemWords(line)
    words1=line.rsplit(",")
    word=words1[0]
    value=words1[1]
    
    #store the sentiment word/value after stemming into the dictionary swords
    #ignore the duplicate word
    if word in swords:
         continue
    else:
        swords[word]=value
#create a dictionary to store sentiment word from twitter text and its frequency

def countwords(results):
   count=dict()
   for value in results["results"]: