<p>In practical application problems, data attributes usually include numerical and categorical attributes, forming mixed attribute data. Due to the inability of existing algorithms to effectively deal with mixed attribute data in clustering problems, this paper proposes an enhanced algorithm for intuitionistic fuzzy clustering of mixed attribute data. First, the proposed algorithm introduces a new distance metric combining Minkowski distance with Value Difference Metric for intuitionistic fuzzy sets, and this metric can handle both numerical and categorical attributes flexibly. Second, a new fitness function is designed to enhance robustness by reducing sensitivity to noisy data, while the differential evolution algorithm is used to enhance the Intuitionistic Fuzzy C-Means algorithm. Finally, experimental results on six UCI datasets demonstrate that the proposed algorithm is effective in terms of accuracy and robustness.</p>