Any sentence of a document has two different aspects: the context around which the sentence is constructed and the grammar using which the sentence is build. Any document is weighted and inferred based on the contextual meaning rather than the grammar used for its construction. For in the context we may have deeply hidden opinions, attitudes, diverse emotions, strong sentiments; may be on business products or on social renderings or on raving Human feelings. These opinions may be healthy and positive or unhealthy and negative. A business may expand its sales line by recommending a product to a customer, if it knows the customer opinion on a product prior. Emotions toned behind the series of social posts if analyzed properly may stop several social issues. In politics, to find the views of the voter, towards a specific political group, in quality assurance by finding errors in the products based on past user experiences and many more application scenarios to visit. For a recommender system sentiment analysis has been proven to be a valuable approach. For digging deeper insights sentiment analysis is using Natural language processing (NLP), Text Mining and Machine Learning techniques. There are many sentiment analysis algorithms, but the predominantly used are the machine learning approaches. Though the rule based approach is less accurate in analyzing the sentiments, the rules which it had defined are base to all the Automatic sentiment analysis algorithms. Automatic sentiment analysis algorithms are machine learning algorithms, which uses the word2vec technique, for digging hidden sentiments. This paper includes a generic survey on the sentiment analysis domain. To reach at an understandable level for the young writers this paper presents a review on how to look into any sentiment analysis based machine learning application.