Improving Chi-Square Feature Selection using a Bernoulli Model for Multi-label Classification of Indonesian-Translated Hadith

Hadith is the foundational knowledge in Islam that must be studied and practiced by Muslims. In Hadith, several types of teachings are beneficial to Muslims all mankind. Some serve as advice, while others contain prohibitions should adhere to. There yet do not belong these categories only information. This study focuses on increasing performance Chi-Square feature selection obtain relevant features for multilabel classification Indonesian-translated Bukhari data. proposes a Chi-Square-based Bernoulli model improve which appropriate short-text data such Hadith. The findings this show proposed method can select based classes; thereby improving with an error value 9.38% compared (9.91%) obtained using basic selection.