senti-wordnet

i need to find the opinion of certain reviews given in websites. i am using sentiwordnet for this. i first send the file containing all the reviews to POS Tagger. tokens=nltk.word_tokenize(line) #tokenization for line in file tagged=nltk.pos_tag(tokens) #for POSTagging print tagged Is there any other accurate way of tokenizing which considers not good as 1 word other than considering it as 2 separate words. Now i have to give postive and negative score to the tokenized words and then calculate the total score. Is there any function in sentiwordnet for this. please help. See First Extract

I plan on using Sentiwordnet 3.0 for Sentiment classification. Could someone clarify as to what the numbers associated with words in Sentiwordnet represent? For e.g. what does 5 in rank#5 mean? Also for POS what is the letter used to represent adverbs? Im assuming 'a' is adjectives. I could not find an explanation either on their site or on other sites. I found the answer. Seems like the number notation comes form Wordnet. It represents the rank in which the given word is commonly used. So rank#5 refers to the context in which rank is used 5th most commonly. Similarly rank#1 refers to the