Sentiment Analysis Is the Process of Understanding the Meaning of Feelings Expressed By an Individual Online About Entities Such As Products, Services, Organizations, Individuals, Issues, Events, Topics, and Interpreting Them In Positive, Negative and Neutral Classes In a Automated Way. the Paper Explains the Workflow of Machine Learning Based Approach Classification For Sentiment Analysis .It Gives an Overview of Common Techniques That Are Used at Different Phases of Classification With Their Strength and Weakness. This Study Would Help In Solving Research Problems That Are Encountered In Sentiment Analysis. the Understanding of the Phases Would Help In Developing Efficient Classifier and Ultimately Improving the Performance of the Classification Algorithm. a Study of Techniques Adopted By Researchers at Different Phases of Sentiment Analysis and Brief Analysis of Study Is Also Presented.