<![CDATA[History of Islamic Culture is one of the contents of Islamic Religious Education subjects that must be taught at Madrasah Ibtidaiyah. Through learning Islamic Cultural History, it is hoped that it can provide motivation for students to know, understand, and live up to Islamic Cultural History which is the basis of their outlook on life. However, some students still consider the subject of Islamic Cultural History to be boring and difficult, because it has extensive/a lot of material, learning is less innovative, so that the interest and motivation of students in studying the history of Islamic culture is low, especially when the learning method used is in the form of lecture. In this study a Question Answering System (QAS) was developed as a learning medium and learning resource that can help students understand the material better. The algorithm used is Conditional Random Field (CRF) which is applied to the Named Entity Recognition process using the Stanford NER library. The test results show a Precision value of 83.4%, Recall of 86.8%, F1-Score 85.6% and overall system Accuracy of 73.97%.Keywords: Question Answering System; Named Entity Recognition; Conditional Random Field; Stanford NER; History of Islamic Culture AbstrakSejarah Kebudayaan Islam merupakan salah satu muatan mata pelajaran Pendidikan Agama Islam yang wajib diajarkan di Madrasah Ibtidaiyah. Melalui pembelajaran Sejarah Kebudayaan Islam, diharapkan dapat memberikan motivasi kepada peserta didik untuk mengenal, memahami, dan menghayati Sejarah Kebudayaan Islam yang menjadi dasar pandangan hidupnya. Namun, sebagian peserta didik masih menganggap mata pelajaran Sejarah Kebudayaan Islam membosankan dan sulit, karena memiliki materi yang luas/banyak, pembelajaran yang kurang inovatif, sehingga minat dan motivasi peserta didik mempelajari sejarah kebudayaan islam menjadi rendah, terutama ketika metode pembelajaran yang digunakan berbentuk metode ceramah. Pada penelitian ini dikembangkan Question Answering System (QAS) sebagai media pembelajaran dan sumber belajar yang dapat membantu siswa memahami materi dengan lebih baik. Algoritme yang digunakan yakni Conditonal Random Field (CRF) yang diterapkan pada proses Named Entity Recognition dengan menggunakan library Stanford NER. Hasil uji menunjukkan nilai Precision sebesar 83.4%, Recall sebesar 86.8%, F1-Score 85.6% dan Accuracy sistem secara keseluruhan sebesar 73.97%.Kata Kunci: Question Answering System; Named Entity Recognition; Conditional Random Field; Stanford NER; Sejarah Kebudayaan Islam]]>
