One of the most common causes of incompleteness is missing data, which occurs when no data value for the variables in observation is stored. These common occurrences significantly affect the conclusions obtained from the specific data. An adaptive approach model outperforming other numerical methods in the classification problem was developed using the class center-based Firefly algorithm. This was accomplished by incorporating attribute correlations into the imputation process (C3FA). However, this model has not been tested on categorical data, which is essential in the preprocessing stage. This is because most machine learning algorithms only consider numerical variables. For processing, encoding is used to convert text or Boolean values into numeric parameters, and the target method is often utilized. This method uses target variable information to encode categorical data. However, it carries the risk of overfitting and inaccuracy within the infrequent categories. Therefore, this study aims to use the smoothing target encoding (STE) method to perform the imputation process by combining C3FA and standard deviation (STD). It also aimed to compare several methods for categorical data, such as mode imputation, DT imputation, and C3FA. The results on the tic tac toe dataset showed that the proposed method (STE + C3FA-STD) produced AUC, CA, F1-Score, precision, and recall values of 0.939, 0.882, 0.881, 0.881, and 0.882, respectively, based on the evaluation using the kNN classifier.