The current paper explores the possibilities of predicting learning styles by applying data classification algorithms. The relevance of the issue of learning styles is important for the smooth running of the educational process. Exploring different methods through which learners acquire knowledge and skills is of immense importance for achieving effective education. For this purpose, we used a suitable dataset containing information about student behavior and their learning styles. The dataset was preprocessed to remove duplicate or empty records. The execution of data classification for predicting learning styles was carried out by creating a process in RapidMiner. We used a large set of classifiers and made a comparative analysis to find specific and maximally accurate results in each category. Based on selected classifiers, a model is proposed for predicting the learning style by applying data classification. The main goal is to achieve personalization of learning by applying a data classification model to predict learning style.