Claim Missing Document
Check
Articles

Found 3 Documents
Search

Analisis Komparatif Metode Bag Of Words, TF-IDF, dan Transformer pada Sistem Penilaian Esai Otomatis Berbasis Kecerdasan Buatan Agustin, Amara Seviany; Widodo, Suprih; Elviani, Ulva; Sari, Ayu Permata; Barri, Muhamad Akda Fathul
AI dan SPK : Jurnal Artificial Intelligent dan Sistem Penunjang Keputusan Vol. 3 No. 2 (2025): Jurnal AI dan SPK : Jurnal Artificial Inteligent dan Sistem Penunjang Keputusan
Publisher : CV. Shofanah Media Berkah

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Penilaian esai secara manual menghadapi kendala inkonsistensi, subjektivitas, dan keterbatasan waktu, terutama pada pembelajaran berskala besar. Penelitian ini membandingkan tiga pendekatan representasi teks pada sistem penilaian esai otomatis berbasis kecerdasan buatan, yaitu Bag of Words (BoW), Term Frequency–Inverse Document Frequency (TF-IDF), dan Transformer (IndoBERT). Dataset yang digunakan berasal dari Kaggle Learning Agency Lab Automated Essay Scoring 2.0 yang terdiri atas 17.207 esai berbahasa Inggris dan diterjemahkan ke bahasa Indonesia menggunakan model Helsinki-NLP opus-mt-en-id. Tahap prapemrosesan meliputi case folding, pembersihan teks, penghapusan stopword, dan stemming menggunakan pustaka Sastrawi. Metode BoW dan TF-IDF dipadukan dengan Support Vector Regression, sedangkan pendekatan Transformer menggunakan fine-tuning IndoBERT. Evaluasi dilakukan menggunakan metrik Quadratic Weighted Kappa (QWK). Hasil eksperimen menunjukkan bahwa IndoBERT mencapai performa tertinggi dengan nilai QWK sebesar 0,7842, diikuti TF-IDF sebesar 0,6521 dan BoW sebesar 0,6103. Meskipun Transformer unggul dari sisi akurasi, metode klasik tetap relevan untuk implementasi dengan keterbatasan sumber daya komputasi karena efisiensi waktu dan kompleksitas yang lebih rendah. Temuan ini menegaskan pentingnya pemilihan metode penilaian otomatis yang disesuaikan dengan konteks kebutuhan dan infrastruktur pendidikan.
Enhancing Computational Thinking in STEM Education through Interactive Programming Media: Evidence of an Equalizing Effect from a Quasi-Experimental Study Widodo, Suprih; Alifah, Aisyah Husna; Septiadi, Jaka; Sari, Dian Permata; Barri, Muhamad Akda Fathul; Maulida, Rahma
Journal of Educational Science and Technology (EST) Volume 11 Number 3 December 2025
Publisher : Universitas Negeri Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26858/est.v11i2.76998

Abstract

Computational Thinking (CT) has increasingly been recognized as a foundational competence in STEM education; however, effective pedagogical strategies that support equitable development of CT skills across students with diverse initial abilities remain underexplored. This study examines the impact of interactive programming media on students’ CT development within a STEM learning context. Employing a quasi-experimental design, two groups of upper secondary students participated in either Code.org–supported instruction or conventional learning. Students’ CT skills were assessed before and after the intervention using a CT-oriented problem-solving instrument.The findings indicate that students who engaged with interactive programming activities demonstrated substantially greater improvement in overall CT performance compared to those in conventional learning environments. Moreover, the results suggest that learning gains were largely independent of students’ initial ability levels, indicating an equalizing effect of interactive programming-based instruction. Improvements were observed across core CT components, including decomposition, abstraction, pattern recognition, and algorithmic thinking. These findings contribute to the theoretical discourse on CT pedagogy by highlighting the role of interactive programming environments as mechanisms that mediate learning equity in STEM education. Pedagogically, the study underscores the potential of block-based programming platforms to support inclusive and conceptually grounded CT learning, particularly in contexts characterized by heterogeneous student readiness.
Vocational Teachers’ Insights on Artificial Intelligence and Computational Thinking Integration Majid, Nuur Wachid Abdul; Barri, Muhamad Akda Fathul; Sari, Ayu Permata; Sodikin, Reisa Aulia; Azman, Mohamed Nor Azhari; Prestoza, Mark Jhon Ramos
Jurnal Penelitian dan Pengkajian Ilmu Pendidikan: e-Saintika Vol. 10 No. 1 (2026): March
Publisher : Lembaga Penelitian dan Pemberdayaan Masyarakat (LITPAM)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36312/e-saintika.v10i1.3660

Abstract

This qualitative case study investigates vocational teachers' perceptions of artificial intelligence and computational thinking integration in Indonesian secondary education. Seven vocational high school teachers from West Java Province participated through semi-structured interviews, non-participant observations, and document analysis following artificial intelligence training. The study employed an interactive analysis model to examine teachers' attitudes, concerns, and implementation strategies regarding artificial intelligence in educational contexts. Findings reveal positive teacher perceptions of artificial intelligence as a learning support tool, with applications ranging from creative media production to technical programming assistance. However, significant ethical and pedagogical concerns emerged, including academic integrity challenges, potential student dependency, inadequate prompt engineering skills, and risks of learning dehumanization. Teachers developed sophisticated guidance strategies, positioning artificial intelligence as a verification tool while maintaining human agency in learning processes. Supporting factors included personal initiative and professional learning communities, while barriers encompassed limited infrastructure, absence of formal guidelines, and varied digital competencies. The research contributes to understanding artificial intelligence adoption challenges in Indonesian vocational education and provides insights for implementing artificial intelligence skills curriculum policy. Results of the research are: (1) educators acknowledge significant advantages of AI integration, especially in improving technical learning experiences and promoting innovative educational endeavors, the integration of AI and computational thinking in the learning process must also prioritize the appropriate pattern aspects and requires clear guidance from stakeholders so that it can be implemented in the learning process; (2)  ethical and pedagogical issues arise as significant obstacles, encompassing abuses of academic integrity, threats of student dependency, insufficient prompt engineering skills, and the potential dehumanization of learning experiences; (3) educators demonstrate significant adaptability by formulating advanced guiding systems that utilize AI as a verification and support mechanism while maintaining human agency in educational decision-making.