Ariesta Kartika Sari
Universitas Trunodjoyo Madura

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

Question Answering Berbasis Transformer pada Koleksi Hadis Kutubut Tis’ah untuk Pembelajaran Pendidikan Agama Islam Ana Tsalitsatun Ni’mah; Riva Ananda Yuanova Kamajaya; Ariesta Kartika Sari
Nuris Journal of Education and Islamic Studies Vol. 6 No. 1: January - June 2026
Publisher : Institut Nurul Islam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52620/jeis.v6i1.229

Abstract

Perkembangan teknologi Natural Language Processing (NLP) berbasis Transformer telah mendorong inovasi dalam sistem Question Answering (QA). Dalam konteks pembelajaran hadis di Pendidikan Agama Islam, akses terhadap koleksi Kutubut Tis’ah masih menghadapi kendala dalam pencarian informasi yang cepat dan kontekstual. Oleh karena itu, diperlukan sistem cerdas yang mampu memahami pertanyaan pengguna dan memberikan jawaban yang relevan secara otomatis. Penelitian ini mengembangkan sistem Question Answering berbasis Transformer dengan memanfaatkan model pra-latih BERT (Bidirectional Encoder Representations from Transformers). Tahapan penelitian meliputi pengumpulan dan praproses data hadis Kutubut Tis’ah, anotasi dataset, perancangan arsitektur model, pelatihan, serta pengujian sistem. Dataset dibagi menjadi data latih dan data uji dengan proporsi 80:20. Evaluasi dilakukan menggunakan metrik Exact Match (EM) dan F1-score untuk mengukur performa model. Hasil pengujian menunjukkan bahwa model yang dikembangkan memperoleh nilai Exact Match (EM) sebesar 78,6% dan F1-score sebesar 86,3%. Sistem mampu memberikan jawaban yang relevan dan kontekstual terhadap berbagai variasi pertanyaan pengguna, serta berhasil mengidentifikasi bagian teks hadis yang sesuai sebagai jawaban dengan tingkat akurasi yang baik. Penerapan model Transformer terbukti efektif dalam meningkatkan kualitas pencarian informasi hadis dibandingkan metode konvensional berbasis pencocokan kata kunci. Sistem ini berpotensi menjadi media pembelajaran hadis yang interaktif, efisien, dan mudah diakses.
Development Of An Augmented Reality-Based Android Educational Game For Learning Computer Networking Materials For Class X Vocational School Students Moh Dani; Ariesta Kartika Sari; Luluk Mauli Diana; Yudha Dwi Putra Negara
EDUTIC Vol 13, No 2: 2026
Publisher : Universitas Trunodjoyo Madura

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21107/edutic.v13i2.35366

Abstract

This research was motivated by the low level of students' motivation and understanding of Computer Network materials, as well as the limited use of interactive learning media at SMK Al-Baisuny. The objectives of this study were to develop an Android-based educational game equipped with Augmented Reality (AR) technology and the Fisher-Yates Shuffle algorithm, and to determine its feasibility, students' learning motivation, and students' understanding after using the developed media. This study employed the Research and Development (RD) method using the ADDIE model, which consists of Analysis, Design, Development, Implementation, and Evaluation stages. Data were collected through expert validation sheets, student response questionnaires, learning motivation questionnaires, and pre-test and post-test instruments. The results showed that the developed media obtained validation scores of 96% from material experts, 96% from test blueprint experts, and 98% from media experts, with an overall average of 96.67%, categorized as highly valid. The trial results indicated student response percentages of 86% in the individual trial, 92% in the small-group trial, and 90% in the large-group trial. Students' learning motivation percentages were 88%, 84%, and 88%, respectively, which were categorized as high. Furthermore, students' understanding based on N-Gain scores reached 92%, 75%, and 59%, indicating that the developed media was able to improve students' understanding. Based on these findings, the Android-based educational game incorporating Augmented Reality technology and the Fisher-Yates Shuffle algorithm is considered feasible for use and can serve as an alternative learning media for Computer Network subjects.