Claim Missing Document
Check
Articles

Found 2 Documents
Search

PENGGUNAAN DEEP LEARNING UNTUK MEMPREDIKSI KINERJA AKADEMIK DAN MEMBERI DUKUNGAN YANG TEPAT BAGI SISWA Wira Andika, Ary; Nurhakim, Lukman; Huzniati Andas, Netty
SIBATIK JOURNAL: Jurnal Ilmiah Bidang Sosial, Ekonomi, Budaya, Teknologi, Dan Pendidikan Vol. 4 No. 7 (2025)
Publisher : Penerbit Lafadz Jaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54443/sibatik.v4i7.3152

Abstract

This research delves into the profound application of deep learning to predict student academic performance and facilitate personalized interventions. By analyzing comprehensive data, including grades, attendance records, and student participation, various deep learning architectures—such as Multilayer Perceptron (MLP), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM)—are employed to uncover subtle patterns indicative of potential learning difficulties. The primary objective of this study is to empower educators and school administrators with predictive insights, enabling them to proactively identify student needs. Targeted interventions, such as personalized academic guidance, relevant emotional support, and appropriate enrichment opportunities, can then be effectively implemented. Nevertheless, this research places crucial emphasis on the careful interpretation of predictive outcomes and vigilance against potential biases within the data. Through the synergy between the analytical power of deep learning and the pedagogical sensitivity of educators, we hope to foster a more inclusive and supportive learning environment, ultimately facilitating the maximum potential development and academic success of every student.
The Affective-Cognitive Advantage of Community Language Learning: A Quasi-Experimental Study on Adolescent EFL Speaking Proficiency and Anxiety Reduction Huzniati Andas, Netty; Karman, Karman; Saputra, Wiwin; Astiantih, Susi; Roslina, Roslina; Andas, Amzal
IJORER : International Journal of Recent Educational Research Vol. 7 No. 5 (2026): September
Publisher : Faculty of Teacher Training and Education Muhammadiyah University of Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46245/ijorer.v7i5.1373

Abstract

Objective: Speaking anxiety remains a major barrier for adolescent EFL learners, particularly in contexts with limited opportunities for authentic communication. Objective: This study compared the efficacy of Community Language Learning (CLL) against the PPP approach in enhancing speaking proficiency and reducing foreign language classroom anxiety among Indonesian junior high school students. Methods: A preregistered, mixed-factorial quasi-experimental design was employed. Eighty eighth-grade students were assigned to CLL (n=40) or PPP (n=40) groups. Speaking proficiency was assessed using a five-component analytic rubric, and anxiety was measured with the Foreign Language Classroom Anxiety Scale (FLCAS), supplemented by reflective diaries. Data were analyzed via mixed-factorial ANOVA and thematic analysis. Result: The CLL group demonstrated significantly greater improvement in overall speaking scores (gain = 27.05) than the PPP group (gain = 19.15), with a large between-groups effect size (d = 1.70). CLL also produced a significantly larger reduction in anxiety (mean reduction = 15.75). Qualitative findings highlighted enhanced psychological safety and collaborative metacognitive correction as key mechanisms. Novelty: This first CLL-PPP comparison in Indonesian junior high schools reveals anxiety reduction as a mediator of speaking gains, integrates mixed-methods evidence, and offers a replicable humanistic EFL model for resource-limited settings.