SoulScripture: Chatbot using Bidirectional Encoder Representations from Transformers as a Medium of Spiritual Guidance

Mental health is an important aspect of human life. Many people face stress, anxiety, and distress daily without adequate support to manage these conditions. Islamic teachings from the Quran and Hadith provide wisdom as a source of inspiration and inner peace. However, accessing and understanding these teachings requires specialized knowledge and often the help of experts. With the advancement of machine learning, these teachings can be made more accessible and accurate. The SoulScripture app offers an innovative solution to support mental health by combining the wisdom of the Quran and Hadith through AI technology. Using the Bidirectional Encoder Representations from Transformers (BERT) model and Transformer architecture, the app can understand and provide relevant advice that anyone can access anytime. This research is significant because it offers a new approach to leveraging technology to support mental well-being, especially for communities underserved by conventional mental health services. The app was developed using self-supervised learning to understand the text of the Hadith without external annotation. This process involves several stages, such as user input, data preprocessing, and text analysis, to generate relevant answers. It is hoped that the SoulScripture application can serve as a source of inspiration and support for individuals in controlling stress and maintaining peace of mind, as well as supporting the achievement of the Sustainable Development Goals (SDGs) related to mental health.