Implementation of a transformer-based question answering model in KutubBot for the Kutubut Tis’ah Hadith Corpus

Current technological developments, particularly in the fields of Artificial Intelligence and Natural Language Processing (NLP), have significantly improved question-answering systems. This technology can be utilized in the fields of education and religion, particularly in the study of hadith. One challenge in hadith learning is the large number of reference sources contained in the Kutubut Tis’ah (nine main hadith books), making it time-consuming for users to find relevant answers. Therefore, a system capable of providing fast, accurate, and appropriate answers based on the hadith text is needed. This research aims to develop an artificial intelligence-based chatbot, called kutubBot, to answer questions related to Kutubut Tis’ah. This system employs a Transformer-based Question Answering approach to comprehend user questions and provide relevant answers based on the hadith corpus. The research stages included data collection from nine hadith books, text preprocessing, model training, and performance evaluation using accuracy, precision, recall, and F1-score metrics. Test results showed that KutubBot achieved good results, with an 97.8% accuracy, 90.3% average precision, 93.45% recall, and an F1-score of 91.86%. These evaluation values indicate that the system has a high level of accuracy and completeness in providing answers that are contextually relevant to the hadith.