Breast cancer remains one of the leading causes of death among women worldwide, primarily due to late detection and diagnosis. Early and accurate prediction is essential to improve survival rates. Machine learning (ML) techniques have proven effective in supporting early diagnosis. This study aims to evaluate and compare the performance of three different approaches: traditional ML, ensemble ML, and deep learning (DL) for early prediction of breast cancer using the Wisconsin Diagnostic Breast Cancer (WDBC) dataset. The methodology includes data collection, preprocessing, and the design of predictive models. Traditional ML algorithms used include Logistic Regression (LR), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Naïve Bayes (NB), and Decision Tree (DT). Ensemble ML techniques comprise Random Forest (RF), XGBoost, and AdaBoost, while DL models include Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN). The models were evaluated using precision, recall, F1-score, and accuracy. The results indicate that XGBoost achieved the highest accuracy (0.99), with strong recall (0.98) and F1-score (0.986), outperforming all other ensemble and traditional ML methods. CNN achieved 0.99 in all evaluation metrics, slightly outperforming RNN, which attained 0.98 accuracy and 0.985 F1-score. These findings confirm that ensemble ML techniques outperform traditional models, while CNN leads among DL models. Furthermore, the proposed models demonstrated superior prediction performance compared to existing studies, particularly in minimizing false negatives, which is critical for healthcare applications.

Purpose: This research aims to address the methodological challenges posed by imbalanced data in classification tasks, where minority classes are severely underrepresented, often leading to biased model performance. It evaluates the effectiveness of hybrid sampling techniques specifically, the Synthetic Minority Oversampling Technique combined with Neighborhood Cleaning Rule (SMOTE-NCL) and with Edited Nearest Neighbors (SMOTE-ENN) in improving the predictive performance of ensemble classifiers, namely Double Random Forest (DRF) and Extremely Randomized Trees (ET), with a focus on enhancing minority class detection.
Methods: A total of eighteen simulated scenarios were developed by varying class imbalance ratios, sample sizes, and feature correlation levels. In addition, empirical data from the 2023 National Socioeconomic Survey (SUSENAS) in Riau Province were employed. The data were partitioned using stratified random sampling (80% training, 20% testing). Models were trained with and without hybrid sampling and optimized through grid search. Their performance was evaluated over 100 iterations using balanced accuracy, sensitivity, and G-mean. Feature importance was interpreted using Shapley Additive Explanations (SHAP).
Results: DRF combined with SMOTE-NCL consistently outperformed all other models, achieving 87.56% balanced accuracy, 82.17% sensitivity, and 86.75% G-mean in the most extreme simulation scenario. On the empirical dataset, the model achieved 76.37% balanced accuracy and 75.49% G-mean.
Novelty: This study introduces a novel integration of hybrid sampling techniques and ensemble learning within an interpretable machine learning framework, providing a robust solution for poverty classification in imbalanced datasets.

Pasar produk perawatan kulit di Indonesia yang sangat kompetitif menuntut inovasi berkelanjutan, terutama pada aspek kemasan yang menjadi titik interaksi pertama dengan konsumen. Kemasan produk yang ada saat ini seringkali memiliki keseragaman visual dan belum sepenuhnya memenuhi harapan fungsional konsumen. Penelitian ini bertujuan untuk mengembangkan desain kemasan sunscreen XYZ yang lebih representatif terhadap preferensi konsumen dengan menggunakan metode Quality Function Deployment (QFD). Pendekatan ini secara sistematis menerjemahkan kebutuhan konsumen, atau Voice of Customer (VoC), yang diperoleh dari analisis Kansei Engineering sebelumnya, ke dalam spesifikasi teknis yang terukur melalui matriks House of Quality (HoQ). Hasil analisis menunjukkan bahwa atribut fungsional seperti 'Praktis', 'Informatif', dan 'Pump' merupakan prioritas utama bagi konsumen. Melalui HoQ, atribut-atribut ini berhasil diterjemahkan menjadi tiga respons teknis dengan prioritas tertinggi: penggunaan botol pipih dengan sudut membulat, penggunaan material HDPE, dan posisi gantungan berada di atas. Berdasarkan spesifikasi tersebut, dikembangkan tiga alternatif desain, di mana Alternatif 3 dengan gaya visual illustrated modern terpilih sebagai konsep terbaik dengan skor kinerja tertinggi. Penelitian ini membuktikan bahwa QFD adalah metode yang efektif untuk memastikan proses pengembangan desain berpusat pada konsumen, menghasilkan konsep kemasan yang unggul secara fungsional dan estetis.

Pertumbuhan jumlah mahasiswa dan pekerja di Kecamatan Kambu, Kota Kendari, mendorong meningkatnya kebutuhan akan hunian sementara seperti kos-kosan. Namun, keterbatasan informasi mengenai lokasi, harga, dan sarana kos menyebabkan kesulitan bagi pencari kos dalam menemukan tempat tinggal yang sesuai. Penelitian ini dilakukan dengan tujuan untuk merancang dan membangun sebuah sistem aplikasi pemetaan kos berbasis web yang menggabungkan teknologi Sistem Informasi Geografis (SIG) guna mempermudah pencarian serta pengelolaan informasi tempat kos. Proses pengembangan sistem mengikuti pendekatan model waterfall, yang meliputi langkah studi pustaka, observasi lapangan, wawancara, analisis kebutuhan sistem, perancangan, implementasi, dan pengujian. Aplikasi dikembangkan dengan memanfaatkan framework Laravel, sementara visualisasi peta dilakukan menggunakan LeafletJS dan pengolahan data spasial menggunakan perangkat lunak QGIS. Hasil yang diperoleh pada penelitian menunjukkan bahwa aplikasi dapat menyajikan informasi kos secara interaktif dalam bentuk peta, memungkinkan pencari kos untuk melihat lokasi, detail harga, fasilitas, serta informasi kontak, dan memudahkan pemilik kos dalam mengelola data kos mereka. Dengan demikian, aplikasi ini memberikan solusi efektif dan efisien dalam memfasilitasi pencarian kos dan pengelolaan hunian di wilayah Kecamatan Kambu.

Web-Based Augmented Reality (AR) is an emerging interactive technology that is increasingly adopted in the fields of visual communication design and digital education. This study aims to analyze trends in the use of Web-Based AR within interactive design contexts and to identify user preferences regarding its features. The research employs a literature review and secondary data analysis, drawing from academic publications, industry reports, and online surveys conducted between 2019 and 2024. The findings reveal a significant increase in the adoption of Web-Based AR both globally and in Indonesia. Key factors driving adoption include interactivity without installation, cross-device compatibility, and rapid browser-based access. Trend diagrams, comparative tables of AR platforms (such as 8thWall, ZapWorks, and WebXR API), and user preference visualizations are presented to support the analysis. The study discusses technological challenges, adoption gaps, and design implications, particularly in education and marketing. Limitations include reliance on secondary data and limited geographic scope. The study’s outcomes are expected to serve as a reference for interactive content developers, educators, and researchers in designing inclusive and adaptive AR-based user experiences.

This study aims to analyze the scientific article writing skills of students at the National Land University. The ability to write scientific articles’s considered important because students are required to be able to express ideas and research findings in a structured, critical, and logical manner. This study adopted quantitative approach with descriptive design. Data were obtained from 51 scientific article assignments of first year students who had completed the Indonesian Language and Writing courses in the Odd Semester of the 2024/2025. The research instrument includes a scientific writing ability assessment rubric, with indicators such as creativity in developing content, use of relevant and credible references, and delivery of knowledge clearly and easily understood. The assessment criteria for the ability to write scientific articles refer to the modified guidelines from Topen which classifies the assessment scores for scientific articles into four groups, namely very good, good, sufficient, and less. The majority of students are at a good level (64.7%) in the creativity of content development, in the ability to use relevant and credible references, the majority of students are also in the good category (52.94%). In the category of conveying knowledge clearly and easily understood, the majority of students also received a good assessment (60.78%). Based on the analysis of students' scientific articles assessed from three main aspects, the students' ability to write scientific articles is classified as good, although in some parts improvements and development are still needed to be even better.

This study aims to describe models of technology integration, namely SAMR and TPACK. Each model explains the components contained therein, which certainly have advantages and disadvantages in their application in PAI learning. The SAMR framework is used to evaluate the level of technology integration in learning, while TPACK helps identify the knowledge required for teachers to effectively combine technology with pedagogy and content. The method used in this study is library research, which involves using relevant literature as a source of information. The findings of this study indicate that: (1) the use of the SAMR model can improve students' critical thinking skills and learning motivation, as well as make the learning process more modern and effective. (2) Although there is significant potential to increase students' interest and teachers' positive perceptions of technology use, challenges such as teachers' readiness and competence, as well as the integration of technology into the curriculum, need to be addressed. (3) Teachers require ongoing training and support to optimise the use of technology in learning, enabling them to effectively apply the SAMR and TPACK models.

This study aims to improve students' learning outcomes in the subject of Civic Education (PKn), particularly on the topic of Human Rights (HAM), which has often posed challenges at the elementary school level due to students’ low comprehension. The Contextual Teaching and Learning (CTL) model was implemented to connect learning materials with students’ real-life experiences. This research employed a Classroom Action Research (CAR) method using the Kemmis and McTaggart model, which includes four stages: planning, action, observation, and reflection, conducted over two cycles. Data were collected through observation, learning outcome tests, and documentation. The data were analyzed using descriptive quantitative analysis to evaluate the improvement in learning outcomes based on the number of students meeting the Minimum Mastery Criteria (KKM). The results showed a significant increase, with 12 out of 17 students (70.59%) achieving the KKM in the first cycle, and 16 students (94.12%) in the second cycle. These findings indicate that the CTL model effectively enhances students’ learning outcomes by helping them relate the concept of human rights to their daily lives. CTL can serve as an innovative alternative teaching strategy for Civic Education at the elementary level. The researcher recommends the continuous application of the CTL model and further development of varied contextual learning materials to achieve more optimal outcomes.

This study focuses on analyzing the cost of capital as a key element in financial decision-making within manufacturing companies in Indonesia. The main objective is to assess the impact of expected return, dividend policy, and systematic risk on the cost of capital, both simultaneously and partially. A quantitative approach was employed using multiple linear regression analysis, with secondary data collected from annual financial reports and market information spanning the years 2020 to 2024. The findings reveal that jointly, the three independent variables significantly influence the cost of capital. However, on a partial basis, only the expected return variable shows a statistically significant and positive effect. In contrast, dividend policy and systematic risk do not exhibit significant influence, and their directional relationships contradict the initial hypotheses. These results suggest that investors’ return expectations play a crucial role in shaping a company’s capital structure, while the effects of dividend policy and systematic risk require further investigation due to their inconsistent outcomes. This study offers practical insights for corporate management and policymakers in formulating effective financing strategies that are responsive to dynamic market conditions.

This study aims to analyze the effectiveness of guided inquiry learning model assisted by animation media on students' learning motivation and high-level thinking skills in geography learning. This study was conducted at SMA Negeri 1 Seririt with subjects of grade XI students. The experimental class and control class were selected using simple random sampling techniques from equivalent pairs based on previous equivalence tests.The research design used was pretest-posttest nonequivalent control group design. The research sample was a pair of equivalent classes, namely class XI 8 consisting of 34 students as the experimental class and class XI 5 consisting of 33 students as the control class. Data collected through observation, tests, questionnaires, documentation. Research data were analyzed descriptively quantitatively. The results of the study showed that: (1) The implementation of the guided inquiry learning model assisted by animation media in geography learning can be applied with very good criteria (88.75), (2) The guided inquiry learning model assisted by animation media is classified as effective in increasing student learning motivation with an experimental class N-Gain index of 0.69> control class N-Gain index of 0.091; and quite effective in increasing students' high-level thinking skills with an N-Gain index of 0.65> control class N-Gain index (0.13). The results of this study have implications for the importance of promoting the application of guided inquiry learning models assisted by animation media to geography teachers to improve students' learning motivation and also students' high-level thinking skills in geography learning.