Training set - proportion of pos / neg / neutral sentences

I am hand tagging twitter messages as Positive, Negative, Neutral. I am try to appreciate is there some logic one can use to identify of the training set what pro开发者_JS百科portion of message should