This paper presents a method of improving fuzzy k-nearest neighbor (k-NN) using genetic algorithm (GA). k-NN is an important classification algorithm. However, a major drawback of the method is that each of the patterns of known classification is considered equally important in the assignment of the pattern to be classified. This can cause difficulties in regions where pattern data overlap. To overcome this drawback, a fuzzy k-NN was proposed for increasing classification correct rate of overlapping pattern data. In this paper, experiments are carried out and illustrate that both k-NN and fuzzy k-NN are still sensitive to the initial k and parameter m. GA is applied to calculate initial k and parameter m. Experimental results show that the our proposed approach has the higher classification correct rate than the fuzzy k-NN.