Abstract:Improving the effectiveness of bread sales requires a data-driven approach capable of identifying historical sales patterns and accurately estimating sales volumes. This study applied the Extreme Gradient Boosting (XGBoost)…
st) algorithm as an ensemble machine learning model to predict bread sales at Syakila Bakery. The data consisted of six months of historical sales data, from July to December 2025, covering sweet bread, white bread, snack bread, and sponge cake products. The research process used the CRISP-DM methodology, consisting of business understanding, data understanding, data preparation, modeling, evaluation, and deployment. The system was developed using PHP as the application interface, Python for the modeling process, and MySQL as the database. Model evaluation was conducted using Mean Absolute Percentage Error (MAPE) and Mean Squared Error (MSE). The evaluation results showed a MAPE of 1.3% and an MSE of 0.21. These results indicate a low prediction error rate, suggesting that the XGBoost model can support more measurable sales planning and facilitate decision-making.
Abstract:Abstract: This study aims to analyze and optimize heart disease risk prediction using Random Forest and XGBoost algorithms combined with feature selection. The dataset used is the Cleveland Heart Disease Dataset, comprising…
ing 303 patient records, 13 predictor features, and a target variable converted into a binary classification. The research stages include preprocessing, data splitting using a stratified train-test split, feature selection, hyperparameter tuning, and evaluation based on accuracy, precision, recall, F1-score, and ROC-AUC. Feature selection yielded eight selected features—thal, cp, thalach, ca, oldpeak, exang, chol, and slope—resulting in a 38.46% reduction in the total number of features. Test results indicate that the Optimized Random Forest model combined with feature selection achieved the best performance, with an accuracy of 91.80%, precision of 89.66%, recall of 92.86%, F1-score of 91.23%, and ROC-AUC of 95.67%.
Keywords: Heart Disease, Random Forest, XGBoost, Feature Selection, Hyperparameter Tuning.
Abstrak: Penelitian ini bertujuan menganalisis dan mengoptimalkan prediksi risiko penyakit jantung menggunakan algoritma Random Forest dan XGBoost berbasis feature selection. Dataset yang digunakan adalah Cleveland Heart Disease Dataset yang terdiri atas 303 data pasien, 13 fitur prediktor, dan satu target yang dikonversi menjadi klasifikasi biner. Tahapan penelitian meliputi preprocessing, pembagian data menggunakan stratified train-test split, feature selection, hyperparameter tuning, serta evaluasi menggunakan accuracy, precision, recall, F1-score, dan ROC-AUC. Feature selection menghasilkan 8 fitur terpilih, yaitu thal, cp, thalach, ca, oldpeak, exang, chol, dan slope, sehingga jumlah fitur berkurang sebesar 38,46%. Hasil pengujian menunjukkan bahwa Optimized Random Forest + Feature Selection memberikan performa terbaik dengan accuracy 91,80%, precision 89,66%, recall 92,86%, F1-score 91,23%, dan ROC-AUC 95,67%.
Kata Kunci: Penyakit Jantung, Random Forest, XGBoost, Feature Selection, Hyperparameter Tuning.
Abstract:Abstract: Digital fraud through short messages has become a national problem in Indonesia; the Indonesia Anti-Scam Centre recorded 637,055 reports with losses of IDR 9.1 trillion up to July 2026. Scam messages are hard to…
o filter with keywords because perpetrators keep changing wording and phone numbers, while official promotional messages share similar linguistic patterns. This study aims to optimize a three-class classifier (normal, fraud, promotion) for Indonesian messages by combining XGBoost, SMOTE, and Bayesian Optimization based on the Tree-structured Parzen Estimator (TPE), and to compare TPE with Grid Search and Random Search under an equal budget of 25 trials. Messages were normalized (URLs, phone numbers, amounts, and slang) and represented with combined word and character n-gram TF-IDF features. On 2,320 messages, XGBoost + SMOTE + TPE achieved a macro F1 of 0.9555, an accuracy of 96.34%, and a fraud-class recall of 0.9507, slightly higher than default XGBoost (F1 0.9531) and Grid Search (F1 0.9529), with a much lighter model that trained 3.2 times faster. Grid Search showed overfitting to the validation folds. The ablation study showed that SMOTE did not help under moderate class imbalance (F1 without SMOTE 0.9581), and Multinomial Naïve Bayes remained the best model (F1 0.9693). Most errors occurred between fraud and promotional messages.
Keywords: Digital Fraud, Text Classification, XGBoost, Bayesian Optimization, SMOTE
Abstrak: Penipuan digital melalui pesan singkat telah menjadi masalah nasional; Indonesia Anti-Scam Centre mencatat 637.055 laporan dengan kerugian Rp9,1 triliun hingga Juli 2026. Pesan penipuan sulit disaring dengan kata kunci karena pelaku terus mengubah variasi kata dan nomor telepon, sementara pesan promosi resmi memiliki pola bahasa yang mirip. Penelitian ini bertujuan mengoptimasi model klasifikasi tiga kelas (normal, penipuan, promo) untuk pesan berbahasa Indonesia dengan menggabungkan XGBoost, SMOTE, dan Bayesian Optimization berbasis Tree-structured Parzen Estimator (TPE), serta membandingkan TPE dengan Grid Search dan Random Search pada anggaran yang sama, yaitu 25 percobaan. Pesan dinormalisasi (tautan, nomor telepon, nominal, dan kata tidak baku) lalu direpresentasikan dengan fitur TF-IDF gabungan n-gram kata dan karakter. Pada 2.320 pesan, XGBoost + SMOTE + TPE mencapai F1-macro 0,9555, akurasi 96,34%, dan recall kelas penipuan 0,9507, sedikit lebih tinggi daripada XGBoost default (F1 0,9531) dan Grid Search (F1 0,9529), dengan model yang jauh lebih ringan dan waktu pelatihan 3,2 kali lebih cepat. Grid Search menunjukkan gejala overfitting terhadap fold validasi. Uji ablasi menunjukkan SMOTE tidak membantu pada ketidakseimbangan kelas moderat (F1 tanpa SMOTE 0,9581), dan Multinomial Naïve Bayes tetap menjadi model terbaik (F1 0,9693). Sebagian besar kesalahan terjadi antara pesan penipuan dan promo.
Kata Kunci: Penipuan Digital, Klasifikasi Teks, XGBoost, Bayesian Optimization, SMOTE.
Abstract:Abstract: Human Resource Management (HRM) is a key element playing a vital role in enhancing productivity, work effectiveness, and organizational sustainability in the era of globalization and digital transformation. The…
increasing complexity of the modern workplace compels organizations to manage human resources that are innovative, adaptive, and focused on competency development. This article aims to analyze the transformation of modern HRM through a competency-based approach, HR management strategies, organizational culture, and the utilization of digital technology to foster sustainable organizational performance. The study employs a literature review method with a qualitative-descriptive approach, examining various scientific articles and academic sources related to human resource management. The findings indicate that HRM strategies encompass workforce planning, recruitment and selection processes, training and skill development programs, and performance management; furthermore, compensation systems significantly influence employee productivity. Additionally, a competency-based approach enables organizations to align personnel with their specific abilities and skills, thereby enhancing work effectiveness and efficiency. A strong and adaptive organizational culture also plays a crucial role in boosting employee motivation, loyalty, and commitment to the organization. Meanwhile, advancements in digital technology drive organizations to implement technology-based HR management systems to improve administrative efficiency and support continuous employee competency development. Thus, the integration of HRM strategies, competencies, organizational culture, and digital transformation serves as a critical factor in shaping efficient, resilient, and sustainable organizations in the contemporary landscape.
Keywords: Human Resource Management, Competency, Organizational Culture, Productivity, Digital Transformation, Sustainable Performance
Abstrak: Manajemen Sumber Daya Manusia (MSDM) adalah salah satu satu elemen kunci yang memiliki peran penting dalam memperbaiki produktivitas, efektivitas kerja, serta keberlanjutan instansi di era globalisasi dan transformasi digital. Perkembangan dunia kerja yang semakin rumit mendorong organisasi untuk dapat mengatur sumber daya manusia yang inovatif, adaptif, dan berorientasi pada pengembangan kompetensi. Artikel ini bertujuan untuk menganalisis transformasi MSDM modern melalui pendekatan berbasis kompetensi, strategi pengelolaan SDM, budaya organisasi, serta pemanfaatan teknologi digital dalam meningkatkan kinerja organisasi yang berkelanjutan. Penelitian ini menggunakan metode penelitian pustaka dengan pendekatan kualitatif deskriptif melalui pengamatan terhadap berbagai artikel ilmiah serta sumber-sumber akademik yang berkaitan dengan pengelolaan sumber daya manusia. Hasil kajian menunjukkan bahwa strategi MSDM mencakup beberapa hal seperti perencanaan tenaga kerja dan proses merekrut dan memilih karyawan, program pelatihan serta pengembangan keterampilan, dan pengelolaan kinerja pegawai. serta sistem kompensasi memiliki pengaruh besar terhadap peningkatan produktivitas kerja karyawan. Selain itu, pendekatan berbasis kompetensi mampu membantu organisasi menempatkan tenaga kerja sesuai kemampuan dan keterampilannya sehingga meningkatkan efektivitas dan efisiensi kerja. Budaya organisasi yang kuat dan adaptif juga berperan penting dalam meningkatkan motivasi, loyalitas, dan komitmen karyawan terhadap organisasi. Di sisi lain, perkembangan teknologi digital mendorong organisasi untuk menerapkan sistem pengelolaan SDM berbasis teknologi guna meningkatkan efisiensi administrasi dan mendukung pengembangan kompetensi karyawan secara berkelanjutan. Dengan demikian, integrasi antara strategi MSDM, kompetensi, budaya organisasi, dan transformasi digital menjadi faktor penting dalam membentuk organisasi yang efisien, tangguh, dan berkelanjutan di masa kini.
Kata Kunci: Manajemen Sumber Daya Manusia, Kompetensi, Budaya Organisasi, Produktivitas, Transformasi Digital, Kinerja Berkelanjutan
Abstract:Eye diseases constitute a major health concern that may lead to visual impairment or blindness if not detected and treated at an early stage. Manual analysis of fundus images requires considerable time, accuracy, and ophthalmological…
thalmological expertise; therefore, a computational method capable of supporting automated and objective classification is needed. This study aims to develop a hybrid transfer learning and XGBoost model for classifying fundus images into four ocular condition categories. The methodology involved applying Contrast Limited Adaptive Histogram Equalization (CLAHE), resizing the images to 224 × 224 pixels, performing retina-specific augmentation on the training data, and stratifying the dataset into training, validation, and testing sets. An ImageNet-pretrained EfficientNetV2S model was employed through transfer learning and fine-tuning to extract image features. The resulting 256-dimensional embeddings, model output probabilities, and statistical image features were subsequently combined, standardized, and classified using XGBoost optimized through five-fold cross-validation. The dataset comprised 4,217 fundus images distributed across four classes: 1,038 cataract images, 1,098 diabetic retinopathy images, 1,007 glaucoma images, and 1,074 normal images, including 2,949 training, 631 validation, and 637 testing images. Evaluation on the 637 test images demonstrated that the proposed transfer learning–XGBoost model achieved an accuracy of 92.78%, macro-precision of 0.93, macro-recall of 0.93, macro F1-score of 0.9265, and one-vs-rest ROC-AUC of 0.9871. The proposed model outperformed the standalone EfficientNetV2S model, which achieved an accuracy of 90.27%, by 2.51 percentage points, with class-specific F1-scores of 0.96 for cataract, 0.99 for diabetic retinopathy, 0.87 for glaucoma, and 0.88 for normal. These findings demonstrate that integrating transfer learning with XGBoost effectively improves eye disease classification and has the potential to support early identification using fundus images, although it is not intended to replace clinical diagnosis by ophthalmologists.
Abstract:Abstract: Student dropout has become a significant challenge for higher education institutions because it negatively affects academic performance, institutional reputation, and resource allocation. Early identification of…
f students at risk of dropping out enables universities to implement timely interventions and improve student retention. This study proposes a predictive model for identifying students at risk of dropping out using the Extreme Gradient Boosting (XGBoost) algorithm based on academic data. The dataset consists of student academic records, including grade point average (GPA), course completion rate, attendance, accumulated credits, failed courses, and semester performance. The data were preprocessed through data cleaning, feature selection, and normalization to improve model performance. The dataset was then divided into training and testing sets using an 80:20 ratio. The XGBoost model was trained and evaluated using accuracy, precision, recall, F1-score, and Area Under the Receiver Operating Characteristic Curve (AUC-ROC). The experimental results demonstrate that XGBoost effectively captures complex relationships among academic variables and provides high predictive performance in identifying students with dropout risk. Feature importance analysis further reveals that GPA, accumulated credits, attendance rate, and the number of failed courses are the most influential factors affecting student dropout. The proposed approach offers a practical decision-support tool for higher education institutions by enabling early detection of at-risk students and facilitating targeted academic support programs. The findings contribute to the application of machine learning in educational data mining and learning analytics, providing valuable insights for improving student retention strategies and academic success. Future research may integrate non-academic factors such as socioeconomic background, psychological characteristics, and student engagement to further enhance prediction accuracy and model generalizability across different educational institutions.
Keywords: Student Dropout Prediction; XGBoost; Machine Learning; Educational Data Mining; Learning Analytics.
Abstrak: Putus kuliah merupakan tantangan signifikan bagi institusi pendidikan tinggi karena berdampak negatif terhadap kinerja akademik, reputasi institusi, dan alokasi sumber daya. Identifikasi dini terhadap mahasiswa yang berisiko putus kuliah memungkinkan universitas untuk menerapkan intervensi yang tepat waktu dan meningkatkan retensi mahasiswa. Penelitian ini mengusulkan model prediktif untuk mengidentifikasi mahasiswa yang berisiko putus kuliah menggunakan algoritma Extreme Gradient Boosting (XGBoost) berdasarkan data akademik. Kumpulan data mencakup rekam jejak akademik mahasiswa, termasuk Indeks Prestasi Kumulatif (IPK), tingkat penyelesaian mata kuliah, kehadiran, akumulasi kredit, mata kuliah yang gagal, dan kinerja semester. Data diproses melalui tahapan pembersihan data, seleksi fitur, dan normalisasi untuk meningkatkan kinerja model. Selanjutnya, kumpulan data dibagi menjadi set pelatihan dan pengujian dengan rasio 80:20. Model XGBoost dilatih dan dievaluasi menggunakan metrik akurasi, presisi, recall, skor-F1, dan Area Under the Receiver Operating Characteristic Curve (AUC-ROC). Hasil eksperimen menunjukkan bahwa XGBoost secara efektif menangkap hubungan kompleks antarvariabel akademik dan memberikan kinerja prediksi yang tinggi dalam mengidentifikasi mahasiswa dengan risiko putus kuliah. Analisis kepentingan fitur mengungkapkan bahwa IPK, akumulasi kredit, tingkat kehadiran, dan jumlah mata kuliah yang gagal merupakan faktor paling berpengaruh terhadap risiko putus kuliah mahasiswa. Pendekatan yang diusulkan ini menawarkan alat pendukung keputusan yang praktis bagi institusi pendidikan tinggi dengan memungkinkan deteksi dini mahasiswa berisiko serta memfasilitasi program dukungan akademik yang terarah. Temuan ini berkontribusi pada penerapan machine learning dalam penambangan data pendidikan dan analitik pembelajaran, serta memberikan wawasan berharga untuk meningkatkan strategi retensi mahasiswa dan keberhasilan akademik. Penelitian di masa mendatang dapat mengintegrasikan faktor non-akademik—seperti latar belakang sosial-ekonomi, karakteristik psikologis, dan keterlibatan mahasiswa—untuk lebih meningkatkan akurasi prediksi dan kemampuan generalisasi model di berbagai institusi pendidikan.
Kata kunci: Prediksi Putus Kuliah Mahasiswa; XGBoost; Machine Learning; Penambangan Data Pendidikan; Analitik Pembelajaran.
Abstract:Abstract: The low interest of Generation Z in saving with Sharia banks poses a challenge to efforts aimed at strengthening the Sharia financial ecosystem in the digital era. As a demographic deeply integrated with technology…
logy and social media, Generation Z requires educational approaches that align with their digital-native characteristics. This study aims to analyze the factors influencing Generation Z's interest in saving with Sharia banks and to formulate a model for boosting this interest through the use of social media and digital education. The study employs a qualitative method with a descriptive approach. Data were collected through in-depth interviews with members of Generation Z and Sharia banking practitioners, Focus Group Discussions (FGDs), and observations of Sharia financial education content on social media. Data analysis was conducted using thematic analysis techniques to identify patterns, perceptions, and experiences regarding Sharia financial literacy and saving behavior. The findings indicate that Sharia financial literacy, digital education, and social media are the primary factors influencing Generation Z's interest in saving. The study proposes an integrative model linking these three factors to enhance saving interest and support the strengthening of the Sharia financial ecosystem in Indonesia.
Keywords: Saving Interest; Generation Z; Islamic Banking; Social Media; Digital Education.
Abstrak: Rendahnya minat menabung Generasi Z pada bank syariah menjadi tantangan dalam upaya penguatan ekosistem keuangan syariah di era digital. Generasi Z sebagai kelompok yang sangat dekat dengan teknologi dan media sosial memerlukan pendekatan edukasi yang sesuai dengan karakteristik digital mereka. Penelitian ini bertujuan untuk menganalisis faktor-faktor yang memengaruhi minat menabung Generasi Z pada bank syariah serta merumuskan model peningkatan minat menabung melalui pemanfaatan media sosial dan edukasi digital. Penelitian menggunakan metode kualitatif dengan pendekatan deskriptif. Data dikumpulkan melalui wawancara mendalam terhadap Generasi Z dan praktisi perbankan syariah, Focus Group Discussion (FGD), serta observasi terhadap konten edukasi keuangan syariah di media sosial. Analisis data dilakukan menggunakan teknik analisis tematik untuk mengidentifikasi pola, persepsi, dan pengalaman informan terkait literasi keuangan syariah dan perilaku menabung. Hasil penelitian menunjukkan bahwa literasi keuangan syariah, edukasi digital, dan media sosial merupakan faktor utama yang memengaruhi minat menabung Generasi Z. Penelitian ini menghasilkan model integratif yang menghubungkan ketiga faktor tersebut dalam meningkatkan minat menabung serta mendukung penguatan ekosistem keuangan syariah di Indonesia.
Kata kunci: Minat Menabung, Generasi Z, Bank Syariah, Media Sosial, Edukasi Digital.
Abstract:Abstract: This study aims to determine the effect of the Teams Games Tournament (TGT) cooperative learning model assisted by interactive Wordwall media on learning motivation and economic learning outcomes of grade XI students…
udents at SMA Negeri 2 Percut Sei Tuan simultaneously and partially. The background of this study is driven by low student learning motivation and a high percentage of economic learning achievements that remain below the Learning Progress Minimum Criteria (KKTP). The research method applied was quantitative with a quasi-experimental approach using the Non-Equivalent Pretest-Posttest Control Group Design. The population comprised all grade XI students at SMA Negeri 2 Percut Sei Tuan, totaling 4 class. The sampling technique used was purposive sampling, which designated class XI-H (36 students) as the experimental class treated with the TGT model assisted by Wordwall media, and class XI-I (36 students) as the control class utilizing the direct learning model. Data on learning motivation were collected via a questionnaire with a Likert scale, while data on learning outcomes were gathered through objective multiple-choice tests. The research instruments were tested for validity and reliability prior to implementation. Data analysis techniques involved descriptive statistics and inferential analysis using Multivariate Analysis of Variance (MANOVA). The results indicated that: (1) There is a positive and significant effect of the TGT learning model assisted by Wordwall media on students' learning motivation and economic learning outcomes simultaneously, proven by the Wilks' Lambda significance value of 0.000 (Sig. < 0.05). (2) There is a positive and significant effect of the TGT learning model assisted by Wordwall media on students' learning motivation partially, indicated by a significance value of 0.000 (Sig. < 0.05). (3) There is a positive and significant effect of the TGT learning model assisted by Wordwall media on students' economic learning outcomes partially, indicated by a significance value of 0.001 (Sig. < 0.05). The features of team competition and digital gamification through Wordwall are proven effective in reducing the tediousness of conceptual material and boosting students' active participation.
Keywords: Teams Games Tournament (TGT); Wordwall; Learning Motivation; Economic Learning Outcomes.
Abstrak: Penelitian ini bertujuan untuk mengetahui pengaruh model pembelajaran kooperatif tipe Teams Games Tournament (TGT) berbantuan media interaktif Wordwall terhadap motivasi belajar dan hasil belajar ekonomi siswa kelas XI SMA Negeri 2 Percut Sei Tuan. Latar belakang penelitian ini didasarkan pada rendahnya motivasi belajar siswa serta banyaknya capaian hasil belajar ekonomi yang masih berada di bawah Kriteria Ketercapaian Tujuan Pembelajaran (KKTP). Metode penelitian yang digunakan adalah kuantitatif dengan pendekatan eksperimen semu (Quasi-Experiment) menggunakan desain Non-Equivalent Pretest-Posttest Control Group Design. Populasi dalam penelitian ini mencakup seluruh siswa kelas XI SMA Negeri 2 Percut Sei Tuan yang mengikuti pelajaran ekonomi sebanyak 4 kelas. Teknik pengambilan sampel menggunakan purposive sampling, yang menetapkan kelas XI-H (36 siswa) sebagai kelas eksperimen dengan perlakuan model TGT berbantuan media Wordwall dan kelas XI-I (36 siswa) sebagai kelas kontrol dengan model pembelajaran langsung. Pengumpulan data motivasi belajar menggunakan instrumen angket dengan skala Likert, sedangkan data hasil belajar dikumpulkan melalui tes obyektif pilihan ganda. Instrumen penelitian telah diuji validitas dan reliabilitasnya sebelum digunakan. Teknik analisis data menggunakan statistik deskriptif dan analisis inferensial Multivariate Analysis of Variance (MANOVA). Hasil penelitian menunjukkan bahwa: (1) Terdapat pengaruh positif dan signifikan model pembelajaran TGT berbantuan media Wordwall terhadap motivasi belajar dan hasil belajar ekonomi siswa secara simultan, dibuktikan dengan nilai signifikansi uji Wilks' Lambda sebesar 0,000 (Sig. < 0,05). (2) Terdapat pengaruh positif dan signifikan model pembelajaran TGT berbantuan media Wordwall terhadap motivasi belajar siswa secara parsial, ditunjukkan oleh nilai signifikansi 0,000 (Sig. < 0,05). (3) Terdapat pengaruh positif dan signifikan model pembelajaran TGT berbantuan media Wordwall terhadap hasil belajar ekonomi siswa secara parsial, ditunjukkan oleh nilai signifikansi 0,001 (Sig. < 0,05). Karakteristik pembelajaran berbasis kompetisi tim dan gamifikasi digital melalui Wordwall terbukti efektif dalam memotong kejenuhan materi konseptual serta meningkatkan keterlibatan aktif siswa.
Kata kunci: Teams Games Tournament (TGT); Wordwall; Motivasi Belajar; Hasil Belajar Ekonomi.
Abstract:Abstract: The advancement of Artificial Intelligence (AI) has driven the widespread adoption of Machine Learning (ML) across various sectors. This study aims to analyze the development trends, implementation sectors, dominant…
inant algorithms, key challenges, and future research directions of ML during the 2020–2026 period. Utilizing a Systematic Literature Review (SLR) based on PRISMA 2020 and Kitchenham Guidelines, 120 peer-reviewed articles from six international academic databases were systematically evaluated. The findings indicate a significant increase in ML publications post-Generative AI era, with the highest adoption rates observed in the healthcare, finance, and manufacturing sectors. Algorithms such as Random Forest, XGBoost, CNN, and LSTM dominate for prediction and classification tasks. Despite its benefits, the primary challenges of ML deployment involve data quality, privacy, algorithmic bias, and model interpretability. This study highlights a shifting research paradigm toward transparent and human-centered AI, outlining future agendas in Explainable AI (XAI), Federated Learning, and Green AI.
Keywords: Artificial Intelligence; Explainable AI; Federated Learning; Generative AI; Machine Learning.
Abstrak: Perkembangan Artificial Intelligence (AI) mendorong pemanfaatan Machine Learning (ML) secara luas di berbagai sektor lintas disiplin. Penelitian ini bertujuan menganalisis tren perkembangan, sektor implementasi, algoritma dominan, tantangan utama, serta arah penelitian masa depan terkait pemanfaatan ML selama periode 2020–2026. Metode yang digunakan adalah Systematic Literature Review (SLR) berbasis pedoman PRISMA 2020 dan Kitchenham Guidelines terhadap 120 artikel ilmiah terpilih dari enam database akademik internasional. Hasil tinjauan menunjukkan adanya peningkatan signifikan publikasi ML pasca-era Generative AI, dengan adopsi tertinggi pada sektor kesehatan, keuangan, dan manufaktur. Algoritma berbasis data tabular (seperti Random Forest dan XGBoost) serta data citra/deret waktu (seperti CNN dan LSTM) menjadi yang paling dominan digunakan untuk kebutuhan prediksi dan klasifikasi. Meskipun menawarkan efisiensi tinggi, tantangan utama implementasi ML berpusat pada kualitas data, privasi, bias algoritma, dan interpretabilitas model. Penelitian ini merumuskan agenda riset masa depan yang mulai bergeser ke arah pengembangan AI yang transparan dan berpusat pada manusia, seperti Explainable AI (XAI), Federated Learning, dan Green AI.
Kata Kunci: Artificial Intelligence; Explainable AI; Federated Learning; Generative AI; Machine Learning.
Abstract:Indonesian language instruction within tertiary educational institutions in Indonesia encounters a significant hurdle due to students' inadequate literacy capabilities. A potential method to address this issue is the integration…
egration of deep learning techniques to enhance language proficiency. This research This research is designed to investigate the implementation of deep learning in the acquisition of the Indonesian language, emphasizing aspects such as historical analysis of the language, spelling verification (EYD), selection of words (diction), arrangement of paragraphs, and scientific writing conventions. The approach employed is a Systematic Literature Review (SLR), aiming to pinpoint and compile existing research pertinent to the use of deep learning in educational settings. Findings indicate that utilizing deep learning can serve as an effective strategy for boosting students' comprehension of the rules governing the Indonesian language through automated feedback that aids in refining spelling, sentence construction, and vocabulary selection. This methodology not only expedites the educational process but also elevates the quality of writing while facilitating the enhancement of students' skills in scientific writing. Deep learning offers a more engaging and relevant framework for studying the Indonesian language at the university level.