Abstract:Abstract: The persistent challenge of low reading motivation and comprehension among English as a Foreign Language (EFL) learners necessitates innovative pedagogical interventions. This study investigates the implementation…
ion of Discovery Learning Method within extensive reading programs as a means to enhance both reading motivation and comprehension among Indonesian university students. Employing a Classroom Action Research (CAR) design conducted over two iterative cycles, the research involved 31 undergraduate students enrolled in an English Class. Data were collected through pre- and post-tests, reading motivation questionnaires, classroom observations, and reflective journals. The intervention integrated Discovery Learning principles—stimulation, problem statement, data collection, data processing, verification, and generalization—into extensive reading sessions where students autonomously selected reading materials and constructed meaning through guided discovery. Quantitative findings revealed substantial improvements across both cycles: mean reading comprehension scores increased from 53.87 (Cycle I) to 73.55 (Cycle II), while reading motivation scores rose from 65.3 to 81.2. Qualitative data indicated enhanced student engagement, increased autonomous learning behaviors, and improved critical thinking skills. The study demonstrates that Discovery Learning Method, when systematically integrated into extensive reading programs, creates a synergistic effect that simultaneously addresses motivational and cognitive dimensions of reading. These findings contribute to the theoretical understanding of constructivist approaches in L2 reading pedagogy and offer practical implications for curriculum designers and English language educators seeking evidence-based strategies to foster meaningful reading engagement in EFL contexts.
Keywords: Discovery Learning, Extensive Reading, Reading Motivation, Reading Comprehension, Classroom Action Research, EFL Pedagogy, Constructivist Learning
Abstrak: Tantangan yang berkelanjutan berupa rendahnya motivasi dan pemahaman membaca pada pembelajar Bahasa Inggris sebagai Bahasa Asing (EFL) menuntut adanya intervensi pedagogis yang inovatif. Penelitian ini menginvestigasi implementasi Metode Discovery Learning dalam program membaca ekstensif sebagai upaya untuk meningkatkan motivasi dan pemahaman membaca mahasiswa perguruan tinggi di Indonesia. Penelitian ini menggunakan desain Penelitian Tindakan Kelas (PTK) yang dilaksanakan dalam dua siklus iteratif dan melibatkan 31 mahasiswa pada kelas bahasa Inggris. Data dikumpulkan melalui pre-test dan post-test, angket motivasi membaca, observasi kelas, serta jurnal reflektif. Intervensi penelitian mengintegrasikan prinsip-prinsip Discovery Learning—stimulasi, perumusan masalah, pengumpulan data, pengolahan data, verifikasi, dan generalisasi—ke dalam sesi membaca ekstensif, di mana mahasiswa secara mandiri memilih bahan bacaan dan mengonstruksi pemahaman melalui proses penemuan yang terarah. Temuan kuantitatif menunjukkan peningkatan yang signifikan pada kedua siklus: rata-rata skor pemahaman membaca meningkat dari 53,87 (Siklus I) menjadi 73,55 (Siklus II), sedangkan skor motivasi membaca meningkat dari 65,3 menjadi 81,2. Data kualitatif mengindikasikan adanya peningkatan keterlibatan mahasiswa, berkembangnya perilaku belajar mandiri, serta peningkatan keterampilan berpikir kritis. Penelitian ini menunjukkan bahwa Metode Discovery Learning, ketika diintegrasikan secara sistematis dalam program membaca ekstensif, menghasilkan efek sinergis yang secara simultan mengakomodasi dimensi motivasional dan kognitif dalam aktivitas membaca. Temuan ini berkontribusi terhadap penguatan pemahaman teoretis mengenai pendekatan konstruktivis dalam pedagogi membaca bahasa kedua serta memberikan implikasi praktis bagi perancang kurikulum dan pendidik bahasa Inggris yang mencari strategi berbasis bukti untuk menumbuhkan keterlibatan membaca yang bermakna dalam konteks EFL.
Kata Kunci: Discovery Learning, Membaca Ekstensif, Motivasi Membaca, Pemahaman Membaca, Penelitian Tindakan Kelas, Pedagogi EFL, Pembelajaran Konstruktivis
Abstract:Abstract: Background: The demands of the workplace in higher education institutions, including polytechnics, require optimal performance from their human resources. Stress, burnout, and work-life imbalance can reduce productivity.…
ductivity. The Employee Well-Being (EWB) program is designed to improve the holistic well-being (physical, mental, social, and financial) of employees. However, the program's effectiveness in boosting performance in the context of Manado State Polytechnic requires empirical study.
Objective: This study aims to analyze the relationship and influence of employee perceptions on the effectiveness of the Employee Well-Being Program on their performance.Method: The study used a quantitative approach with a survey method. A sample of 50 employees (lecturers and educational staff) at Manado State Polytechnic was selected using a purposive sampling technique. Data were collected through a closed questionnaire measuring two variables: (1) Perception of the Effectiveness of the EWB Program (24 indicators including physical health, mental support, social relationships, and financial well-being) and (2) Employee Performance (10 task-based and contextual indicators). Data were analyzed statistically using the Pearson Product Moment correlation test and simple linear regression analysis.Results: The analysis results show a correlation coefficient (r) of 0.712, which indicates a strong positive relationship between the perception of the effectiveness of the EWB Program and employee performance. The R Square value (coefficient of determination) is 0.507, meaning that approximately 50.7% of the variation in employee performance can be explained by variations in their perceptions of the effectiveness of the EWB Program. The results of the regression test show the equation Y = 15.245 + 0.789X, with a significance value (p)
Abstract:Abstract: Machine learning (ML) models are frequently characterized as "black boxes" due to their complexity, which renders them difficult for humans to interpret. Model interpretability is crucial for understanding the…
underlying drivers of specific predictions. In the context of mining operations, explaining the fuel consumption rate (FCR) patterns of mining haul trucks through predictive modeling is essential; however, engineers often struggle to identify the most significant contributors quickly and easily. Because standard ML models do not disclose the logic behind their decisions, engineers face ambiguity when analyzing conditions and prioritizing necessary repairs. Such prioritization is vital, as maintenance costs, technical difficulty, and downtime directly impact productivity. Consequently, a model-agnostic approach is required to bridge this gap. This research aims to develop a predictive model to analyze FCR behavior and patterns, subsequently interpreting them through model-agnostic techniques. The study utilized Vehicle Health Monitoring System (VHMS) data from August 2024 to February 2025, incorporating outlier and multicollinearity management. The Random Forest Regressor (RFR) was employed as the primary machine learning algorithm. Global interpretations were conducted using Partial Dependence Plots (PDP), Feature Interaction, and Permutation Feature Importance, while local interpretations were performed using Local Interpretable Model-Agnostic Explanations (LIME) and Shapley Values (SHAP). The performance evaluation results demonstrate that the RFR predictive model maintains consistent performance regardless of the data treatment applied. The optimal configuration was the RFR model without normalization or outlier handling, achieving an RMSE of 3.7312, a SMAPE of 4.64%, and an R-squared of 0.7936. Both global and local interpretations identified engine speed, road angle, and boost pressure as the top three factors significantly contributing to FCR.
Keywords: Fuel Consumption Rate; Interpretable Machine Learning; Agnostic Model; Random Forest
Abstrak: Model machine learning (ML) sering disebut sebagai “Black-Box†karena kerumitannya sehingga sulit diinterpretasikan oleh manusia. Interpretabilitas model menjadi sangat penting untuk memahami penyebab sebuah prediksi tertentu dibuat. Salah satunya dalam memahami perilaku dan menjelaskan pola fuel consumption rate (FCR) dari mining haul truck menggunakan model prediksi. Seorang engineer akan kesulitan untuk menentukan kontributor paling signifikan secara mudah dan cepat. Pada sebuah prediksi, sebuah model ML tidak akan memberi tahu bagaimana sampai pada sebuah keputusan. Hal ini akan menimbulkan kebingungan engineer pada saat akan menganalisa kondisi dan menentukan prioritas perbaikan yang diperlukan. Prioritisasi perbaikan perlu dilakukan karena pertimbangan biaya, tingkat kesulitan, dan downtime yang sangat mempengaruhi produktivitas. Oleh karena itu, pendekatan model agnostik perlu dilakukan. Penelitian ini bertujuan untuk menghasilkan model prediksi untuk memahami perilaku dan pola FCR kemudian menginterpretasikannya dengan model agnostik. Penelitian ini menggunakan data Vehicle Health Monitoring System (VHMS) dari Agustus 2024 hingga Februari 2025 dengan penanganan outlier dan multikolinieritas. Algoritma ML yang digunakan adalah Random Forest Regressor (RFR). Model agnostik yang digunakan untuk interpretasi global adalah Partial Dependence Plot (PDP), Feature Interaction, dan Permutation Feature Importance. Sedangkan interpretasi lokal menggunakan Local Interpretable Model-Agnostic Explanations (LIME) dan Shapley Value (SHAP). Hasil evaluasi performa model menunjukkan bahwa model prediksi RFR memiliki performa yang konsisten bagaimanapun perlakuan data diterapkan. Model prediksi terbaik yang dipilih adalah model RFR Tanpa Normalisasi – Tanpa Penanganan Outlier dengan nilai RMSE 3,7312, SMAPE 4,64%, dan R-Squared 0,7936. Hasil interpretasi global dan lokal menunjukkan bahwa top three faktor yang berkontribusi signifikan terhadap FCR adalah engine speed, road angle, dan boost pressure.
Kata Kunci: Fuel Consumption Rate; Interpretable Machine Learning; Model Agnostik; Random Forest
Abstract:Abstract: High levels of stress among university students are a critical issue that can affect mental health, well-being, and academic performance. This study aims to develop a real-time student stress detection system using…
sing physiological data integrated with IoT technology and machine learning algorithms. The data used includes body temperature, blood oxygen saturation (SpOâ‚‚), heart rate, and blood pressure, acquired via embedded sensors and automatically transmitted to the cloud. The classification model was built using a combination of Random Forest and XGBoost, with enhanced accuracy through SMOTE-based data balancing and hyperparameter optimization using Optuna. The system was tested on a dataset of 3,420 records, classified into four stress levels: anxious, calm, tense, and relaxed. Evaluation results showed that the Random Forest model achieved the highest accuracy of 91%, followed by RF + XGBoost and RF + XGBoost + Optuna with accuracies of 90% each. The final model was deployed in a user interface using Streamlit, allowing real-time stress classification from IoT sensor input and manual input testing. The system proved to be effective and responsive in detecting stress objectively and can support digital-based mental health monitoring and counseling services for students.
Keywords: Stress detection, IoT, Machine Learning, Random Forest, XGBoost
Abstrak: Tingkat stres yang tinggi di kalangan mahasiswa merupakan permasalahan serius yang dapat memengaruhi kesehatan mental, kesejahteraan, dan performa akademik. Penelitian ini bertujuan untuk mengembangkan sistem deteksi tingkat stres mahasiswa secara real-time menggunakan data fisiologis berbasis teknologi IoT dan algoritma machine learning. Data yang digunakan meliputi suhu tubuh, kadar oksigen dalam darah (SpOâ‚‚), detak jantung, dan tekanan darah yang diperoleh melalui sensor terintegrasi dan dikirim ke cloud secara otomatis. Model klasifikasi yang dikembangkan memanfaatkan kombinasi algoritma Random Forest dan XGBoost, dengan peningkatan akurasi melalui teknik balancing data menggunakan SMOTE dan optimasi hyperparameter otomatis menggunakan Optuna. Sistem diuji menggunakan dataset berjumlah 3.420 data dengan distribusi empat kelas stres: cemas, tenang, tegang, dan rileks. Hasil evaluasi menunjukkan bahwa model Random Forest menghasilkan akurasi tertinggi sebesar 91%, disusul oleh RF + XGBoost dan RF + XGBoost + Optuna dengan akurasi masing-masing sebesar 90%. Model akhir kemudian diintegrasikan ke dalam antarmuka pengguna berbasis Streamlit, yang memungkinkan klasifikasi stres secara real-time dari data sensor IoT dan juga melalui input manual. Sistem ini terbukti efektif dan responsif dalam mendeteksi stres secara objektif dan dapat digunakan untuk mendukung layanan konseling atau pemantauan kesehatan mental mahasiswa secara digital.
Kata kunci: Deteksi stres, IoT, Machine Learning, Random Forest, XGBoost
Abstract:Abstract: This study aims to examine how marketing mix components influence consumer purchasing decisions at Show Koffie Cafe. The research employs a quantitative methodology utilizing survey techniques, where information…
n was gathered through questionnaires distributed to cafe patrons. A total of 100 participants were involved in this investigation, representing diverse age groups, genders, and income levels. Multiple linear regression analysis was applied to determine the impact of product, pricing, promotional activities, and location variables on purchase decisions. The analytical findings reveal that all marketing mix variables demonstrate positive and significant effects on purchasing decisions, with promotional strategies emerging as the most influential factor. These discoveries provide strategic insights for Show Koffie Cafe management to develop more effective marketing initiatives aimed at attracting additional customers and boosting sales performance. Furthermore, this research emphasizes the critical importance of coordinated integration among marketing mix elements to achieve optimal influence on consumer behavior patterns.
Â
Keyword: Purchase Decision, Product, Price, Promotion, and Place
Â
Abstrak: Penelitian ini bertujuan untuk mengkaji bagaimana komponen bauran pemasaran mempengaruhi keputusan pembelian konsumen di Kafe Show Koffie. Penelitian ini menggunakan metodologi kuantitatif dengan memanfaatkan teknik survei, dimana informasi dikumpulkan melalui kuesioner yang disebarkan kepada pengunjung kafe. Sebanyak 100 partisipan terlibat dalam penyelidikan ini, mewakili kelompok usia, jenis kelamin, dan tingkat pendapatan yang beragam. Analisis regresi linier berganda diterapkan untuk menentukan dampak variabel produk, penetapan harga, aktivitas promosi, dan lokasi terhadap keputusan pembelian. Temuan analitis mengungkapkan bahwa seluruh variabel bauran pemasaran menunjukkan efek positif dan signifikan terhadap keputusan pembelian, dengan strategi promosi muncul sebagai faktor yang paling berpengaruh. Penemuan ini memberikan wawasan strategis bagi manajemen Kafe Show Koffie untuk mengembangkan inisiatif pemasaran yang lebih efektif yang ditujukan untuk menarik pelanggan tambahan dan meningkatkan kinerja penjualan. Selanjutnya, penelitian ini menekankan pentingnya kritis dari integrasi terkoordinasi di antara elemen-elemen bauran pemasaran untuk mencapai pengaruh optimal pada pola perilaku konsumen.
Â
Kata kunci: Keputusan Pembelian, Produk, Harga, Promosi Dan Tempat
Â
Abstract:Abstract: This study examines the effect of sales promotion on consumer purchase intention in Lazada e-commerce. A quantitative method was applied with a questionnaire to 100 active users in Medan. Data were analyzed…
zed using simple linear regression. Results indicate that sales promotion—including discounts, vouchers, and cashback—has a positive and significant impact on purchase intention (β = 0.686; p < 0.05). The regression equation shows every unit increase in promotion raises purchase intention by 0.423. Effective promotional strategies are therefore essential to boost online sales and consumer engagement.
Â
Keywords: sales promotion; purchase intention; e-commerce; Lazada; marketing strategy
Â
Abstrak: Penelitian ini menganalisis pengaruh promosi penjualan terhadap minat beli konsumen pada platform e-commerce Lazada. Metode kuantitatif digunakan dengan penyebaran kuesioner kepada 100 pengguna aktif di Medan dan dianalisis menggunakan regresi linear sederhana. Hasil menunjukkan promosi penjualan—diskon, voucher, dan cashback—berpengaruh positif serta signifikan terhadap minat beli (β = 0,686; p < 0,05). Persamaan menegaskan bahwa peningkatan promosi meningkatkan minat beli. Temuan ini menekankan pentingnya strategi promosi efektif untuk mendorong penjualan dan keterlibatan konsumen.
Â
Kata kunci: promosi penjualan; minat beli; e-commerce; Lazada; strategi pemasaran
Abstract:Abstract: Cars are a very popular four-wheeled means of transportation today, so that many consumers or buyers are interested in buying new or used cars depending on their respective economies. One factor that influences…
consumer interest in buying a car is price. Price greatly influences the sustainability of consumers in buying a car. It is necessary to estimate the estimated price of a used car based on criteria such as mileage, taxes, fuel consumption, and engine capacity. Estimation using regression method where in regression method there are 7 more methods including (1) Linear Regression, (2) Support Vector Regression – Linear, (3) Support Vector Regression – RBF, (4) Decision Tree Regression, (5) Random Forest Regressor, (6) Gradient Boosting Regression, (7) NLP Regressor applied in this research. Based on the 7 regression methods, the best method with the best accuracy value will be sought which will be used in the deployment processing process to determine the price of used cars with a ratio of 90:10, 80:20 and 70:30 producing the best estimated value is decision tree regression. Each method has a high level of accuracy including in the 90:10 ratio decision tree regression as the best method in the ratio has an accuracy level of 99%, and in the 80:20 ratio decision tree regression has an accuracy value of 99%, then the 70:30 ratio decision tree regression again becomes the best method with an accuracy level of 99%.
Â
Keyword: Machine Learning; Used Car Price Prediction; Regression; Performance Model
Â
Abstrak: Mobil merupakan sebuah alat transportasi kendaraan roda empat yang sangat populer saat ini, sehingga banyak sekali minat konsumen atau pembeli yang ingin membeli mobil baru maupun bekas tergantung dari ekonomi nya masing-masing Salah satu yang mempengaruhi minat konsumen dalam membeli mobil yaitu harga. Harga sangat berpengaruh dalam keberlangsungan konsumen dalam membeli suatu mobil Perlunya upaya estimasi untuk mengetahui perkiraan harga mobil bekas dengan berdasarkan kriteria seperti jarak tempuh, pajak, konsumsi bahan bakar, serta kapasitas mesin. Estimasi menggunakan metode regresi dimana dalam metode regresi terdapat 7 metode lagi meliputi (1) Linear Regression, (2) Support Vector Regression – Linear, (3) Support Vector Regression – RBF, (4) Decision Tree Regression, (5) Random Forest Regressor, (6) Gradient Boosting Regression, (7) NLP Regressor yang diterapkan dalam penelitian ini. Berdasarkan 7 metode regresi tersebut akan dicari 1 metode terbaik dengan nilai akurasi paling terbaik yang akan digunakan dalam proses pengolahan deploy untuk menentukan harga mobil bekas dengan rasio yaitu 90:10, 80:20 dan 70:30 menghasilkan nilai estimasi terbaik adalah decision tree regression. Masing – masing metode memiliki tingkat akurasi yang tinggi diantaranya dalam rasio 90:10 decision tree regression sebagai metode terbaik dalam rasio tersebut memiliki tingkat akurasi sebesar 99%, dan pada rasio 80:20 decision tree regression tersebut memiliki nilai akurasi sebesar 99%, selanjutnya rasio 70:30 decision tree regression kembali menjadi metode terbaik dengan tingkat akurasi sebesar 99%.
Â
Kata kunci: Machine Learning; Prediksi Harga Mobil Bekas; Regresi; Performance Model
Abstract:Abstract: The Palestinian conflict has become a global issue that has triggered significant public responses, especially through social media platforms. This study aims to evaluate the performance of a hybrid Naïve Bayes–XGBoost…
yes–XGBoost model in classifying netizen sentiment toward the Free Palestine issue on platform X (formerly Twitter) in the year 2025. Data were collected using the X (Twitter) API with keywords such as #freepalestine and #savegaza, then processed through a series of preprocessing stages and sentiment labeling using a lexicon-based approach. The dataset was then split into 80% training data and 20% testing data to compare the performance of the baseline Naïve Bayes model and the hybrid model. The evaluation results show that the baseline Naïve Bayes model achieved an accuracy of 75.7% and an F1-score of 76%, while the hybrid Naïve Bayes–XGBoost model achieved a significantly higher accuracy of 95.5% and an F1-score of 96%. These findings indicate that integrating the two algorithms improves both accuracy and balance in sentiment classification, especially for unstructured and imbalanced data. This study recommends the use of hybrid models for public opinion analysis on social media and suggests further development using deep learning approaches.
Â
Keywords: Sentiment Analysis, Free Palestine, Naïve Bayes, XGBoost, Hybrid Model
Abstrak: Konflik Palestina menjadi isu global yang memicu respons besar dari masyarakat dunia, terutama melalui media sosial. Penelitian ini bertujuan untuk mengevaluasi performa model hybrid Naïve Bayes–XGBoost dalam mengklasifikasikan sentimen netizen terhadap isu Free Palestine di platform X (Twitter) tahun 2025. Data dikumpulkan menggunakan X (Twitter) API dengan kata kunci #freepalestine dan #savegaza, lalu diproses melalui tahapan preprocessing, dan pelabelan menggunakan pendekatan lexicon-based. Selanjutnya, data dibagi menjadi data latih (80%) dan data uji (20%) untuk membandingkan performa antara model Naïve Bayes dasar dan model hybrid. Hasil evaluasi menunjukkan bahwa model Naïve Bayes dasar menghasilkan akurasi 75,7% dan F1-score 76%, sedangkan model hybrid Naïve Bayes–XGBoost mencapai akurasi 95,5% dan F1-score 96%. Temuan ini menunjukkan bahwa integrasi kedua algoritma mampu meningkatkan akurasi dan keseimbangan klasifikasi sentimen, khususnya pada data yang tidak terstruktur dan imbalanced. Penelitian ini merekomendasikan penggunaan model hybrid untuk analisis opini publik dimedia sosial, serta pengembangan lebih lanjut menggunakan pendekatan deep learning.
Â
Kata kunci: Sentimen, Free Palestine, Naïve Bayes, XGBoost, Hybrid Model
Abstract:Abstract: Language style is a way of speaking/writing that depends on the situation of the person (or people) you are speaking or writing to. Language style refers to the selection of linguistic forms to convey social or…
artistic effects. In communication, people usually use formal or informal language, depending on the situation they are communicating with others. Language style also tells the listener to accept what is being said, whether it is serious, ironic, funny, or otherwise (Saputri et al., 2022). According to (Joos Theory Author et al., 2023) in (Joos in Chaer and Agustina (1995: 92)) divides variety of language into five styles, they are; frozen style, formal style, consultative style, casual style, and intimate style. This study investigates how the role of language style in TikTok live streaming in selling products on the platform with qualitative methods. Qualitative is used to describe and analyse the phenomena that usually occur in daily activities and does not deal with numbers as a data (Saputri et al., 2022). In collecting data, videos were taken from three tiktok accounts, namely the Elformula Official, Nalalacollection, and FJ.bottle accounts with a duration of 10 minutes each. language style plays a critical role in the success of live-streaming sales on TikTok by fostering effective communication and engaging interaction between sellers and consumers. The casual and intimate language styles dominate this environment, creating a friendly and relatable atmosphere that appeals to TikTok's diverse audience. Sellers use dynamic, easy-to-understand, and personalized language to build trust and establish a connection with their viewers, enhancing customer engagement and loyalty.
Â
Keywords: Language style; Semantics; Social Media; Tiktok.
Abstract:Penerapan Algoritma Adaboost dalam mengetahui pola pengguna KB di Puskesmas Tanjung Harapan yang dapat membantu memberikan informasi kepada pengguna KB terkait alat KB yang sebaiknya digunakan dan juga dapat mempermudah…
pendataan pengguna KB di ruang lingkup Puskesmas Tanjung Harapan. Penerapan Algoritma Adaboost dalam mengetahui pola pengguna KB di Puskesmas Tangjung Harapan, dibuat menggunakan bahasa pemrograman Visual Basic .Net. Berdasarkan pengujian blackbox yang telah dilakukan, didapatkan hasil bahwa fungsionalitas dari aplikasi sudah berjalan dengan baik dan penerapan Algoritma Adaboost dalam mengetahui pola pengguna KB di Puskesmas Tanjung Harapan telah sesuai dengan yang diharapkan.