One of the superior capabilities of fuzzy systems is that they can use the information expressed in a linguistic pattern. Though most fuzzy systems, have been formed to emulate human decision making behaviour, the linguistic information stated by an expert may not be precise or that it is difficult for the expert to articulate the accumulated knowledge to encompass all circumstances. Hence, it is essential to provide a learning capability for fuzzy systems, namely, to generate or modify the expert rules based on experiences. In this paper, a review of techniques available for updating/learning the parameters of a fuzzy system are presented.