Analisis Level Pemahaman Al-Qur’an Menggunakan Kecerdasan Buatan Berdasarkan Teori Taksonomi Bloom:

Sebuah Solusi atas Tantangan Objektivitas Penilaian dalam Pendidikan Islam

Abstract

This study explores the role of artificial intelligence (AI)  with Natural Language Processing (NLP) language management  in collaboration with Machine Learning (ML) and Deep Learning (DL) which produces modern language modeling techniques such as Word Embeddings, N-Gram, Recurrent Neural Networks (RNN), Transformer Models, Generative Pre-trainde Transformer (GPT) and its variants, such as Bidirectional Encoder Representations from Transformers (BERT) and Long Short-Term Memory (LSTM) in carrying out Quranic understanding activities. The success of AI with its NLP has been proven to be able to understand the Quran well. However, there are several challenges and limitations that cannot be done while still requiring the role of humans because they are dealing with the text of the Qur'an in classical Arabic with its morphology, syntax and semantic complexity. As well as concerns that algorithms may not be able to understand the cultural, historical, and theological nuances that often affect the interpretation of the verse. Therefore, it is appropriate to conduct research to measure the level of understanding of the Quran carried out by AI using the Bloom taxonomy instrument. Using a qualitative approach and literature study related to AI features with its NLP, Bloom's taxonomy combines the two. The results of this study show that AI capabilities when measured with Bloom's taxonomy are fully achieved at level 1, namely Remembering, while at the level of Understanding, Applying, Analyzing, the ability is limited. And not able to do it at the level of Evaluating and Creating.