Human interact with natural language. Words are uncertain, vague and imprecise. Besides, words mean different things to different people. So, the use of numerical model to represent words is a difficult task. In this paper, we propose to give a new review of the most used computing with words methods especially encoding ones. Many approaches were used for computing with words. Some of these methods convert words into a representation using fuzzy logic like the person MF approach, the interval endpoint approach, the interval approach and the enhanced interval approach. These methods were used in the encoding part of computing with words (encoding: how to transform a word into a fuzzy set)