“…The goal was to prepare data or text obtained from unstructured social media into good data and could be easily processed for further processing (Jimenez-Marquez et al, 2019;Shu et al, 2017;Iskandar & Marjuki, 2022). In the preprocessing technique there were several processes such as parsing, case folding, tokenizing, stemming, filtering/stop words, normalization (Chen et al, 2020;Baccouche et al, 2020;Sarimole & Fadillah, 2022). In the text normalization stage, it was very important to be able to help parse Indonesian language that could understand lexical meaning well, performance in processing structural and unstructured words could be improved if the preprocessing stages were carried out properly, especially normalization for unstructured words (Izonin et al, 2022;Basan et al, 2022).…”