Abstract:Family communication plays a fundamental role in shaping an individual's character, particularly in adulthood. This study examines the influence of family communication patterns on adult character formation through a bibliometric…
liometric analysis. Utilizing a Systematic Literature Review (SLR) and bibliometric approach, this research explores key trends, theoretical frameworks, and scholarly discussions related to family communication and character development. The study highlights the significance of open, trust-based, and emotionally supportive communication in fostering essential adult traits such as responsibility, integrity, empathy, and adaptability. Findings suggest that families with high conversation orientation cultivate individuals with stronger social skills and problem-solving abilities, whereas rigid conformity-oriented communication may hinder independent decision-making. Additionally, digital communication emerges as a growing factor influencing family dynamics. By mapping research trends and analyzing citation patterns, this study provides valuable insights into the long-term impact of family communication on adulthood and offers recommendations for future research.
Abstract:Research on leadership styles and the strengthening of school culture in Islamic schools has become increasingly significant in response to the growing demand for educational quality grounded in religious values. School…
principals play a strategic role in developing an adaptive, collaborative, and character-oriented organizational culture that supports students’ holistic development. This study focuses on publication trends, thematic cluster structures, and the tendencies of leadership style implementation in strengthening Islamic school culture based on bibliometric analysis. The study employed a descriptive quantitative approach using bibliometric analysis of 397 Scopus-indexed documents published between 2019-2025. Data collection was conducted using Publish or Perish, while data visualization analysis was performed using VOSviewer through keyword co-occurrence techniques. The findings reveal that school culture and distributed leadership emerged as the most dominant themes and occupied central positions within the bibliometric network. The shift in research paradigms indicates a transition from structural approaches toward participatory, collaborative, and contextual approaches in modern educational leadership. Another finding demonstrates that the concepts of Islamic school and Islamic leadership have not yet appeared prominently in international research networks, thereby indicating a significant research gap regarding the integration of Islamic values into school culture and educational leadership practices. This study concludes that the strengthening of school culture is influenced not only by the administrative capabilities of school principals, but also by their ability to foster collaboration, social relations, innovation, and the continuous internalization of moral values. The novelty of this research lies in the use of bibliometric analysis to systematically map the relationship between leadership styles and Islamic school culture, while also offering a perspective on integrating the values of amanah, shura, and exemplary conduct into modern collaborative leadership models.
Abstract:The cognitive development of early childhood requires appropriate stimulation, one of which is through color recognition. Color block media serves as an educational tool that not only introduces various colors but also familiarizes…
amiliarizes children with geometric shapes, numerical concepts, and trains their thinking and memory skills. This study aims to implement color block media as an innovative learning method to effectively improve early childhood abilities in color recognition. Early childhood is a stage of exploration, where the learning process must be concrete, engaging, and enjoyable. Color block media combines visual and manipulative approaches that can foster curiosity and active involvement in the learning process. This learning-through-play activity encourages children to naturally identify, differentiate, and name colors. The study used a quantitative method with a pretest-posttest design to determine the effectiveness of the media. The research was conducted at KB Adduriyah 3 on October 29, 2024, and data analysis was performed using a t-test through SPSS 18 for Windows. The results showed a significance value (2-tailed) of 0.00. Since this value is smaller than the significance level (α = 0.05), the null hypothesis is rejected, and the alternative hypothesis is accepted. The results of the study demonstrate that color block media has a significant impact on improving children's color recognition skills. Additionally, the media also enhances children's active participation in learning activities. Therefore, color block media is highly recommended as a creative and effective learning strategy for educators and parents in supporting the cognitive development of early childhood.
Abstract:This study, conducted from 2017 to 2023, analyzes the relationship between international tourism and law using bibliometric methods on a dataset of 237 selected documents. The findings reveal a declining trend in annual…
dataset growth, with an average document age of approximately 2.93 years. The study delves into author collaboration, document types, and keyword usage. There was an average of 4,042 citations per document, reflecting the scholarly impact in the field. The dataset included 714 plus keywords and 738 author keywords, contributing to content analysis. Among the 570 unique authors, 87 were sole authors, while 99 documents were single-authored, and the average co-authorship was 2.47. Notably, there was no international co-authorship in the dataset. Document types encompassed articles, books, book chapters, conference papers, and more, with peak scientific production observed in 2020. Treemap and thematic map analysis visualize term distribution and research trends.
Abstract:Abstract: This study aims to classify the nutritional status of toddlers based on anthropometric data using the K-Nearest Neighbor (KNN) algorithm. Data were obtained from 20 Integrated Health Posts (Posyandu) in Rumbai…
Timur District, including Lembah Sari Village and Limbungan Village with a total of 1,000 toddler data. After cleaning and preprocessing, 782 data were obtained ready for use. The preprocessing stages include data cleaning and transformation, outlier removal, minority class handling, and data normalization. Next, data balancing was carried out using the Synthetic Minority Oversampling Technique (SMOTE) to address class imbalance. The data was divided into 80% training data and 20% test data, then the K parameter was tested from 1 to 15 using 5-fold cross-validation. The results showed that the value of K = 1 provided the best performance with a macro recall of 0.8827 and an accuracy of 86.26%. These results indicate that the combination of the KNN algorithm with the SMOTE method and Min-Max normalization is effective in improving classification performance on imbalanced data and producing accurate and balanced predictions of toddler nutritional status between classes.
Keywords: k-nearest neighbor; toddler nutritional status; SMOTE; min-max scaling; classification; anthropometric data
Abstrak: Penelitian ini bertujuan untuk mengklasifikasikan status gizi balita berdasarkan data antropometri menggunakan algoritma K-Nearest Neighbor (KNN). Data diperoleh dari 20 Posyandu di Kecamatan Rumbai Timur, meliputi Kelurahan Lembah Sari dan Kelurahan Limbungan dengan total 1.000 data balita. Setelah melalui proses cleaning dan preprocessing, diperoleh 782 data yang siap digunakan. Tahapan pra-pemrosesan meliputi pembersihan dan transformasi data, penghapusan outlier, penanganan kelas minoritas, serta normalisasi data. Selanjutnya dilakukan penyeimbangan data menggunakan Synthetic Minority Oversampling Technique (SMOTE) untuk mengatasi ketidakseimbangan kelas. Data dibagi menjadi 80% data latih dan 20% data uji, kemudian dilakukan pengujian parameter K dari 1 hingga 15 menggunakan 5-fold cross-validation. Hasil penelitian menunjukkan bahwa nilai K = 1 memberikan performa terbaik dengan recall macro sebesar 0,8827 dan akurasi 86,26%. Hasil ini menunjukkan bahwa kombinasi algoritma KNN dengan metode SMOTE dan normalisasi Min-Max efektif dalam meningkatkan kinerja klasifikasi pada data tidak seimbang serta menghasilkan prediksi status gizi balita yang akurat dan seimbang antar kelas.
Kata kunci: k-nearest neighbor; status gizi balita; SMOTE; min-max scaling; klasifikasi; data antropometri
Abstract:This study focuses on the development of an Augmented Reality (AR)–based learning application designed to assist students in understanding the mathematical concepts of volume and surface area of three-dimensional geometric…
tric shapes. The development process adopted the Multimedia Development Life Cycle (MDLC) model, which consists of six systematic stages: concept, design, material collecting, assembly, testing, and distribution. The research concentrated on the development and expert validation stages. Validation results from content and media experts indicate that the application meets pedagogical and technical feasibility standards. The content expert confirmed that the materials align with the national mathematics curriculum and are presented in a clear, contextual, and accurate manner, while the media expert highlighted the user-friendly interface, interactive features, and visual appeal of
the application. Theoretically, this AR-based medium bridges the gap between abstract mathematical concepts and concrete visualization by enabling students to interact directly with
virtual 3D objects. Practically, the application enhances learning motivation and engagement by providing dynamic, interactive experiences. Overall, this research contributes to the
advancement of educational technology by offering a systematic model for developing AR-based learning media that support active and meaningful learning in the digital
era.
Abstract:Abstract: Rehabilitation programs are essential in correctional systems to equip inmates with the skills and behavioral readiness required for social reintegration. However, rehabilitation program assignment in many correctional…
ectional institutions remains dependent on manual and subjective assessments, which may result in inconsistent decisions. This study develops a Random Forest–based prediction system to support objective and data-driven rehabilitation program determination. A quantitative approach was applied using historical inmate data from January 2023 to January 2025, comprising 2,023 records. The research process included data preprocessing, an 80:20 training–testing split, model training, and performance evaluation using accuracy, precision, recall, and F1-score metrics. The results show that the model achieved an accuracy of 86.17% during training in Google Colab and 68.83% when deployed within the application system. This performance gap reflects real-world deployment and computational constraints rather than model failure. The proposed system provides consistent and objective rehabilitation program recommendations, thereby supporting more effective rehabilitation planning and decision-making in correctional institutions.
Keywords: correctional institutions; inmate rehabilitation programs; machine learning; random Forest; prediction system
Abstrak: Program pembinaan narapidana memiliki peran penting dalam sistem pemasyarakatan untuk membekali warga binaan dengan keterampilan serta kesiapan perilaku dalam proses reintegrasi ke masyarakat. Namun, pada banyak lembaga pemasyarakatan, penentuan program pembinaan masih bergantung pada penilaian manual yang bersifat subjektif, sehingga berpotensi menimbulkan ketidakkonsistenan dalam pengambilan keputusan. Penelitian ini mengembangkan sistem prediksi program pembinaan narapidana berbasis algoritma Random Forest guna mendukung pengambilan keputusan yang objektif dan berbasis data. Pendekatan kuantitatif diterapkan menggunakan data historis narapidana periode Januari 2023 hingga Januari 2025 sebanyak 2.023 data. Tahapan penelitian meliputi prapemrosesan data, pembagian data latih dan uji dengan rasio 80:20, pelatihan model, serta evaluasi performa menggunakan metrik akurasi, precision, recall, dan F1-score. Hasil penelitian menunjukkan bahwa model mencapai akurasi sebesar 86,17% pada tahap pelatihan di Google Colab dan 68,83% saat diimplementasikan pada sistem aplikasi. Perbedaan performa tersebut mencerminkan keterbatasan lingkungan operasional, bukan kegagalan model. Secara keseluruhan, sistem yang dikembangkan mampu memberikan rekomendasi program pembinaan yang lebih objektif dan konsisten, sehingga mendukung perencanaan pembinaan yang lebih efektif.
Kata kunci: mesin pembelajaran; program pembinaan narapidana; random Forest; sistem pemasyarakatan; sistem prediksi
Abstract:YouTube has become a major platform for public discourse in Indonesia, yet large-scale sentiment analysis of its comments remains challenging due to dynamic content, informal language, and limited labeled data. This study…
y proposes a Selenium–IndoBERT pipeline for sentiment analysis of Indonesian YouTube comments using a pseudo-labeling approach. Data were collected from ten YouTube videos discussing the One Piece flag phenomenon, yielding 10,842 comments after preprocessing. Selenium was employed to extract comments from dynamic pages, while IndoBERT was fine-tuned on a small manually labeled dataset and used to generate pseudo-labels for unlabeled data. Model performance was evaluated using probabilistic metrics, including Coverage, Expected Calibration Error (ECE), and Brier Score. At a confidence threshold of 0.75, 78.5% of comments received pseudo-labels, with an ECE of 0.095 and a Brier Score of 0.174. Manual validation showed substantial agreement with human annotations (Fleiss’ kappa = 0.72). The results indicate that the proposed pipeline enables scalable and reliable sentiment analysis with minimal manual annotation.
Abstract:Abstract: This study discusses the optimization of RAG for a FAQ system in the field of information technology product security certification at BSSN. Although LLM generate reliable responses, they often lack up-to-date…
and domain-specific knowledge, which can be addressed through the RAG approach. This research aims to optimize a domain-specific RAG system by improving embedding performance, enhancing prompt robustness, and increasing retrieval accuracy. The research methods consist of three stages. The first stage involves fine-tuning the bge-m3 embedding model and evaluating its performance using MRR, Recall, and AUC. The second stage applies prompt engineering techniques, namely the SRSM and Autodefense, to mitigate direct-injection and escape-character prompt injection attacks. The third stage evaluates the proposed RAG system using Precision, Recall, and F1-Score metrics against four baseline models. The results of research show that the fine-tuned embedding model achieves higher performance than the original model, with MRR@1 and Recall@1 values of 0.80 and an AUC@100 of 0.7023. In addition, the proposed prompt engineering techniques demonstrate robustness against prompt injection attacks, while the overall RAG system attains a perfect Precision, Recall, and F1-Score of 1.00. In conclusion, the proposed approach effectively enhances retrieval accuracy, embedding quality, and system security, resulting in a more reliable RAG-based FAQ system for information technology product security certification.
Keywords: embedding fine-tuning; large language model; prompt engineering; prompt injection mitigation; retrieval-augmented generation
Abstrak: Studi ini membahas optimasi RAG untuk sistem FAQ di bidang sertifikasi keamanan produk teknologi informasi di BSSN. Meskipun LLM menghasilkan respons yang andal, mereka seringkali kurang memiliki pengetahuan terkini dan spesifik domain, yang dapat diatasi melalui pendekatan RAG. Penelitian ini bertujuan untuk mengoptimalkan sistem RAG spesifik domain dengan meningkatkan kinerja embedding, meningkatkan ketahanan prompt dan meningkatkan akurasi pengambilan. Metode penelitian terdiri dari tiga tahap. Tahap pertama melibatkan fine-tuning model embedding bge-m3 dan mengevaluasi kinerjanya menggunakan Mean Reciprocal Rank (MRR), Recall, dan AUC. Tahap kedua menerapkan teknik rekayasa prompt, yaitu Self- SRSM dan Autodefense, untuk mengurangi serangan direct-injection dan escape-character prompt injection. Tahap ketiga mengevaluasi sistem RAG yang diusulkan menggunakan metrik Presisi, Recall, dan F1-Score terhadap empat model dasar. Hasil penelitian menunjukkan bahwa model embedding yang disempurnakan mencapai kinerja yang lebih tinggi daripada model asli, dengan nilai MRR@1 dan Recall@1 sebesar 0,80 dan AUC@100 sebesar 0,7023. Selain itu, teknik rekayasa prompt yang diusulkan menunjukkan ketahanan terhadap serangan injeksi prompt, sementara sistem RAG secara keseluruhan mencapai Presisi, Recall, dan F1-Score sempurna sebesar 1,00. Kesimpulannya, pendekatan yang diusulkan secara efektif meningkatkan akurasi pengambilan, kualitas embedding dan keamanan sistem, menghasilkan sistem FAQ berbasis RAG yang lebih andal untuk sertifikasi keamanan produk teknologi informasi.
Kata kunci: penyempurnaan embedding; model bahasa besar; rekayasa prompt; mitigasi injeksi prompt; retrieval-augmented generation
Abstract:Abstract: One of the main challenges in digital image processing is limited resolution, which makes it difficult to preserve visual details when images are enlarged. Conventional methods such as Bilinear Interpolation are…
e commonly used for image upscaling; however, these approaches often produce blurred images, lose fine textures, and fail to reconstruct complex visual structures. This study aims to enhance digital image resolution by employing a deep learni based approach using a Low-Light Convolutional Neural Network (LLCNN) built upon a Deep Neural Network (DNN) architecture. The dataset used in this study is the DIV2K dataset, which consists of 1,000 high-resolution images. These images were downsampled using scaling factors of ×2, ×3, and ×4 to generate paired Low Resolution–High Resolution (LR–HR) data for training and evaluation. The proposed LLCNN is designed to extract important features such as edges, textures, and local patterns through multiple convolutional layers, followed by non-linear mapping to reconstruct high-resolution images more accurately. Quantitative performance evaluation was conducted using the Peak Signal-to-Noise Ratio (PSNR) and the Structural Similarity Index (SSIM). Model performance was evaluated quantitatively using the Peak Signal-to-Noise Ratio (PSNR) metric. Experimental results showed that the proposed method improved image quality compared to the bilinear method. These results indicate that the deep learning based approach effectively improves image sharpness and structural fidelity, thereby demonstrating its potential for digital image resolution enhancement.
Keywords: deep neural network; image resolution; low-light convolutional neural network; machine learning
Abstrak: Permasalahan utama dalam pengolahan citra digital adalah keterbatasan resolusi yang menyebabkan detail visual sulit dipertahankan ketika citra diperbesar. Metode konvensional seperti Bilinear Interpolation masih banyak digunakan, namun sering menghasilkan citra buram, kehilangan tekstur halus, serta tidak mampu merekonstruksi struktur visual yang kompleks. Penelitian ini bertujuan untuk meningkatkan kualitas resolusi citra digital dengan memanfaatkan pendekatan deep learning berbasis Low-Light Convolutional Neural Network (LLCNN) yang dibangun di atas arsitektur Deep Neural Network (DNN). Data yang digunakan dalam penelitian ini berasal dari dataset DIV2K, yang terdiri dari 1000 citra beresolusi tinggi. Citra tersebut diturunkan menjadi resolusi rendah menggunakan faktor downsampling ×2, ×3, dan ×4 untuk membentuk pasangan data Low Resolution–High Resolution (LR–HR) sebagai data pelatihan dan pengujian. LLCNN dirancang untuk mengekstraksi fitur-fitur penting seperti tepi, tekstur, dan pola lokal melalui beberapa lapisan konvolusi, kemudian melakukan pemetaan non-linear guna merekonstruksi citra resolusi tinggi secara lebih presisi. Evaluasi performa model dilakukan secara kuantitatif menggunakan metrik Peak Signal-to-Noise Ratio (PSNR). Hasil eksperimen menunjukkan bahwa metode yang diusulkan mampu meningkatkan kualitas citra dibandingkan metode bilinear. Hasil ini membuktikan bahwa pendekatan berbasis deep learning efektif dalam meningkatkan ketajaman dan kesesuaian struktur citra digital.
Kata kunci: deep neural network; low-light convolutional neural network; machine learning; resolusi citra