Internet and WWW are growing a huge amount of data from the users entered the text that is related to the product review, opinions, attitudes, and any other services. This information is processing and analyzing various tools to be used. The NLP and Information retrieval methods are used to analyze the information and understanding these process information. The main problem of Sentiment analysis is classified into two categories that are Positive opinion or negative opinion. This paper focus to compare the lexical based sentiment classification and Machine learning based sentiment classification methods. Various algorithms and approaches and features are classified in this paper. Different methods are taken to solve the sentiment analysis problems.