AbstractIn present times a great number of clustering algorithms are available which group objects having similar features. But most of the datasets have data values that are categorical, which makes it difficult to implement these algorithms. The concept of genetic algorithm on intuitionistic fuzzy k-Mode method is proposed in the paper to cluster categorical data. This model is an extension of intuitionistic fuzzy k-Mode in which the notion of fitness related objective functions, crossovers, mutations and probability has been added to provide better clusters for the data objects. Also the intuitionistic parameter has been retained for the calculation of membership values of element x in a given cluster. UCI repository datasets were used for assessing efficacy of algorithms. The qualified analysis and results depict much consistent performance, where a significant improvement is achieved as compared to intuitionistic fuzzy k-Mode and simulated annealing based intuitionistic fuzzy k-mode. Genetic Algorithm based intuitionistic fuzzy k-Mode is very efficient when clustering is applied on large datasets that are categorical in nature, which proves to be very critical for data mining processes.