Abstract:The advancement of information technology has encouraged companies to manage data more effectively in decision-making processes, including inventory management. Problems such as overstock, deadstock, and stockout frequently…
tly occur due to the lack of data-driven management based on historical transactions. This study aims to analyze and classify inventory demand levels using the Naïve Bayes algorithm to optimize inventory management at PT Soala Gogo Natama. This research employs a quantitative descriptive method with a case study approach. The dataset consists of 300 observations divided into training data (210) and testing data (90) with 9 variables. The analysis process includes data preprocessing, implementation of the Gaussian Naïve Bayes algorithm using Python, and model evaluation using accuracy, confusion matrix, and classification report. The results indicate that the model successfully classifies data into three categories: low, medium, and high, with an accuracy of 78.89%. Model evaluation shows good performance in the low and high classes, while misclassification occurs in the medium class due to overlapping characteristics between classes. This suggests that the model is effective in identifying data patterns but has limitations when handling data with similar distributions. In conclusion, the Naïve Bayes algorithm is effective in supporting inventory management decision-making, particularly in determining stock priorities based on historical data.
Abstract:Text based social engineering attacks are a growing cyber threat that is difficult to detect by conventional intrusion detection systems, especially in previously unobserved or zero-day variants. This study proposes a Natural…
tural Language Processing Open-Set Intrusion Detection System (NLP-OSIDS) framework that integrates Term Frequency-Inverse Document Frequency (TF-IDF) trigram (1.3-gram) feature representation with an Open-Set Multilayer Perceptron architecture based on energy based scoring to detect zero-day social engineering attacks without requiring training examples from that class. Experiments were conducted on the public dataset phishing_email.csv with 82,486 combined samples from Enron, SpamAssassin, Nazario, Ling, CEAS, and Nigerian Fraud datasets with strict zero-day partitioning following open-set recognition evaluation standards. The results show that NLP-OSIDS achieved an AUROC of 0.7808, surpassing all closed-set baselines (AUROC = 0.500) with the lowest False Positive Rate of 0.0088, while the Zero-Day Detection Rate (ZD-DR) of 0.077 indicates the need for adaptive threshold optimization as a direction for further research.
Abstract:Abstract:This study discusses the effect of Yttrium (Y) doping variations on the physical and electromagnetic properties of Barium Hexafere (BaFe??O??) synthesized using the solidstate reaction method. The doping variations…
ons used were 0.00; 0.01; 0.03; and 0.05 at%. The calcination process was carried out at 600 °C and sintering at 1200 °C. Characterization was performed using X-Ray Diffraction (XRD), Scanning Electron Microscopy (SEM), and Vector Network Analyzer (VNA). The XRD results showed that all samples had a single phase of BaFe??O?? with a hexagonal structure according to JCPDS No. 84-0757 without secondary phases, with the highest crystallinity at a doping of 0.05 at%. SEM analysis showed an increase in grain density and a decrease in porosity with increasing Y doping. VNA testing showed the best Reflection Loss value of –12.98 dB at a frequency of 8.87 GHz with 71.8% microwave absorption. Yttrium doping of 0.05 at% is the optimal composition for enhancing microwave absorption in Barium Hexafere.
Keywords:Barium Hexafere, Yttrium, solid state reaction, XRD, SEM, VNA
Abstrak: Penelitian ini membahas pengaruh variasi doping Yttrium (Y) terhadap sifat fisis dan elektromagnetik Barium Heksaferrit (BaFe??O??) yang disintesis menggunakan metode solid state reaction. Variasi doping yang digunakan yaitu 0,00; 0,01; 0,03; dan 0,05 at%. Proses kalsinasi dilakukan pada suhu 600 °C dan sintering pada suhu 1200 °C. Karakterisasi dilakukan menggunakan X-Ray Diffraction (XRD), Scanning Electron Microscopy (SEM), dan Vector Network Analyzer (VNA). Hasil XRD menunjukkan semua sampel memiliki fasa tunggal BaFe??O?? dengan struktur heksagonal sesuai JCPDS No. 84-0757 tanpa fasa sekunder, dengan kristalinitas tertinggi pada doping 0,05 at%. Analisis SEM menunjukkan peningkatan kerapatan butir dan penurunan porositas seiring bertambahnya doping Y. Uji VNA menunjukkan nilai Reflection Loss terbaik –12,98 dB pada frekuensi 8,87 GHz dengan penyerapan gelombang mikro 71,8%. Doping Yttrium 0,05 at% merupakan komposisi optimal untuk meningkatkan kemampuan penyerapan gelombang mikro pada Barium Heksaferrit.
Kata kunci: Barium Heksaferrit, Yttrium, solid state reaction, XRD, SEM, VNA
Abstract:Abstract: The quality of healthcare services is heavily dependent on the effective management of medical equipment. At Maddin Batu Bara Clinic, logistics management currently relies on conventional physical documentation.…
. This manual approach triggers various operational hurdles, such as difficulties in real-time stock monitoring, delayed procurement, and a high risk of both stockouts and overstocking. Furthermore, coordination with suppliers remains unintegrated, and the clinic lacks a standardized calculation method to determine reorder timing.This study aims to evaluate medical equipment management and existing supply chain constraints, subsequently designing a web-based Electronic Supply Chain Management (E-SCM) system. The system integrates the Reorder Point (ROP) method to optimize the availability of medical supplies. The research utilizes an applied method with a descriptive approach and quantitative analysis, with data collected through observation, interviews, and documentation.The results indicate that the implementation of the E-SCM system using the ROP method accurately determines reorder points based on average consumption rates and delivery lead times. This digital innovation is proven to enhance inventory management efficiency, minimize the risk of stock imbalances, and strengthen communication synergy between the clinic and suppliers within a more modern supply chain ecosystem.
Keywords: E-SCM; Medical Equipment; Availability; Reorder Point (ROP); Maddin Clinic.
Abstrak: Kualitas pelayanan kesehatan sangat bergantung pada efektivitas pengelolaan alat medis. Di Klinik Maddin Batu Bara, manajemen logistik saat ini masih mengandalkan pencatatan fisik secara konvensional. Metode yang mengandalkan dokumen fisik ini memicu berbagai hambatan operasional, seperti sulitnya memantau stok secara real-time, keterlambatan pengadaan barang, hingga tingginya risiko kekurangan (stockout) maupun kelebihan persediaan (overstock). Selain itu, koordinasi dengan pemasok belum terintegrasi dan klinik belum memiliki standar perhitungan untuk menentukan waktu pemesanan kembali. Penelitian ini dilakukan untuk mengevaluasi manajemen alat medis dan hambatan rantai pasok yang ada, kemudian merancang solusi berupa sistem Electronic Supply Chain Management (E-SCM) berbasis web. Sistem ini mengintegrasikan metode Reorder Point (ROP) untuk mengoptimalkan ketersediaan alat medis. Metode penelitian yang diterapkan adalah penelitian terapan dengan pendekatan deskriptif dan analisis kuantitatif melalui teknik observasi, wawancara, serta dokumentasi. Hasil penelitian menunjukkan bahwa implementasi sistem E-SCM dengan metode ROP mampu menentukan titik pemesanan ulang secara akurat berdasarkan tingkat konsumsi rata-rata dan lead time pengiriman. Inovasi digital ini terbukti meningkatkan efisiensi manajemen inventaris, meminimalkan risiko ketidakseimbangan stok, serta memperkuat sinergi komunikasi antara pihak klinik dan pemasok dalam ekosistem rantai pasok yang lebih modern.
Kata kunci: E-SCM; Alat Medis; Ketersediaan; Reorder Point (ROP); Klinik Maddin
Abstract:Abstract: In the era of Industrial Revolution 4.0 and the transition toward Society 5.0, Indonesia's marine and fisheries sector faces a digital transformation imperative to maintain global competitiveness and resource sustainability.…
ustainability. Fisheries extension officers, acting as agents of change, are required to possess adaptive digital competencies to facilitate technology adoption among stakeholders. However, existing training programs are often generic and lack prioritization based on real-world needs and strategic impact. This study aims to map digital competency gaps and formulate strategic training priorities by integrating individual needs with expert perspectives. The research employs a quantitative approach with a two-stage analytical design. A case study was conducted at the Base Administrative Unit (Satminkal) of the Center for Fisheries Training and Extension (BPPP) Medan. In the first stage, a Training Needs Analysis (TNA) was conducted on 230 extension officers to measure digital competency gaps across 12 indicators, followed by a Pareto Analysis to identify critical competencies (the vital few). In the second stage, the five critical competencies identified from the TNA were evaluated by a panel of nine experts using the Analytical Hierarchy Process (AHP) to determine final priorities based on four criteria: urgency, impact on performance, policy alignment, and gap severity. TNA findings revealed the highest gaps in technical competencies: "Operation of Fisheries-Specific Applications" (gap -1.33) and "Digital Content Production" (gap -1.31). Pareto Analysis confirmed that 78.4% of the total gap originated from five vital competencies. AHP results indicated that "Impact on Performance" was the primary determinant (weight 40.1%). The synthesis produced "Operation of Fisheries-Specific Applications" as the top priority (global weight 0.397). The integration of TNA and AHP produces more precise training priorities for effective resource allocation in the digital age.
Keywords: Fisheries Extension Officers; Digital Competence; Training Needs Analysis; Pareto Analysis; Analytical Hierarchy Process
Abstrak: Di era Revolusi Industri 4.0 dan transisi menuju Society 5.0, sektor kelautan dan perikanan Indonesia menghadapi imperatif transformasi digital guna mempertahankan daya saing global dan keberlanjutan sumber daya. Penyuluh perikanan, sebagai agen perubahan, dituntut memiliki kompetensi digital yang adaptif untuk memfasilitasi adopsi teknologi di kalangan pelaku usaha. Namun, program pelatihan yang ada seringkali bersifat umum dan tidak diprioritaskan berdasarkan kebutuhan riil dan dampak strategisnya. Penelitian ini bertujuan untuk memetakan kesenjangan kompetensi digital penyuluh perikanan dan merumuskan prioritas pelatihan yang strategis dengan mengintegrasikan perspektif kebutuhan individu dan pertimbangan ahli. Penelitian menggunakan pendekatan kuantitatif dengan desain analisis bertahap (two-stage quantitative analysis). Studi kasus dilakukan di Satminkal Balai Pelatihan dan Penyuluhan Perikanan (BPPP) Medan. Pada tahap pertama, Training Needs Analysis (TNA) dilakukan terhadap 230 penyuluh untuk mengukur kesenjangan kompetensi digital pada 12 indikator, dilanjutkan dengan Analisis Pareto untuk mengidentifikasi kompetensi kritis (vital few). Pada tahap kedua, lima kompetensi kritis dievaluasi oleh 9 panel ahli menggunakan metode Analytical Hierarchy Process (AHP) untuk menentukan prioritas akhir berdasarkan empat kriteria: urgensi, dampak kinerja, kesesuaian kebijakan, dan tingkat kesenjangan. Hasil TNA mengungkap kesenjangan tertinggi pada "Pengoperasian Aplikasi Spesifik" (gap -1,33) dan "Produksi Konten Digital" (gap -1,31). Analisis Pareto mengonfirmasi 78,4% total kesenjangan berasal dari lima kompetensi vital. Hasil AHP menunjukkan kriteria "Dampak pada Kinerja" menjadi determinan utama (bobot 40,1%). Sintesis prioritas global menghasilkan "Pengoperasian Aplikasi Spesifik" sebagai prioritas utama (bobot 0,397). Integrasi TNA dan AHP menghasilkan prioritas pelatihan yang lebih presisi untuk efektivitas alokasi sumber daya di era digital.
Kata Kunci: Penyuluh Perikanan; Kompetensi Digital; Training Needs Analysis; Analisis Pareto; Analytical Hierarchy Process
Abstract:Abstract: Hospital Occupational Safety and Health (OSH) is a crucial aspect in ensuring the safety of healthcare workers, patients, and the hospital work environment. The implementation of OSH is a particular concern, particularly…
rticularly in hospitals in Padang City, which face high and complex occupational risks. This study aims to systematically examine the implementation of OSH in hospitals in Padang City through a literature review. The method used was a search of scientific articles from various databases such as Google Scholar, PubMed, and ScienceDirect, with the inclusion criteria being articles published within the last five years and relevant to the topic of OSH. The study results indicate that OSH implementation in hospitals in Padang City has been ongoing but is not yet optimal, characterized by limitations in policies, human resources, compliance with standard operating procedures, and a lack of training and supervision. Furthermore, safety culture and management commitment also influence the success of OSH implementation. The study concludes that improving OSH implementation requires strengthening internal regulations, improving workforce competency, and a sustained commitment from hospital management to create a safe and healthy work environment.
Keywords: K3RS, Occupational Safety, Hospital, Implementation, Literature Review.
Abstrak: Keselamatan dan Kesehatan Kerja Rumah Sakit (K3RS) merupakan salah satu aspek penting dalam menjamin keselamatan tenaga kesehatan, pasien, serta lingkungan kerja di rumah sakit. Implementasi K3RS menjadi perhatian khusus, terutama di rumah sakit di Kota Padang yang memiliki risiko kerja tinggi dan kompleks. Penelitian ini bertujuan untuk mengkaji secara sistematis implementasi K3RS di rumah sakit di Kota Padang melalui pendekatan literature review. Metode yang digunakan adalah penelusuran artikel ilmiah dari berbagai database seperti Google Scholar, PubMed, dan ScienceDirect dengan kriteria inklusi artikel yang diterbitkan dalam 5 tahun terakhir dan relevan dengan topik K3RS. Hasil kajian menunjukkan bahwa implementasi K3RS di rumah sakit di Kota Padang telah berjalan namun belum optimal, ditandai dengan masih adanya keterbatasan dalam kebijakan, sumber daya manusia, kepatuhan terhadap standar operasional prosedur, serta kurangnya pelatihan dan pengawasan. Selain itu, faktor budaya keselamatan dan komitmen manajemen juga berpengaruh terhadap keberhasilan implementasi K3RS. Kesimpulan dari studi ini adalah bahwa peningkatan implementasi K3RS memerlukan penguatan regulasi internal, peningkatan kompetensi tenaga kerja, serta komitmen berkelanjutan dari pihak manajemen rumah sakit guna menciptakan lingkungan kerja yang aman dan sehat.
Kata Kunci: K3RS, Keselamatan Kerja, Rumah Sakit, Implementasi, Literature Review.
Abstract:Abstract: Hospitals are high-risk work environments, involving exposure to biological, chemical, physical, ergonomic, and psychosocial hazards that can impact worker safety. Therefore, the implementation of Occupational…
Safety and Health (OHS) in Human Resources (HR) management is a crucial aspect in preventing occupational risks. This study aims to analyze the implementation of OHS in HR management in hospitals as a preventive strategy against workplace accidents and occupational diseases. The method used was a literature review with a qualitative descriptive approach, through a review of various national scientific articles published between 2022 and 2025, available in PDF format. The study results indicate that effective OHS implementation is influenced by several key factors, including management commitment, HR training and competency, the availability of infrastructure, and a work safety culture within the hospital environment. Furthermore, the integration of OHS into HR management functions such as recruitment, training, performance appraisal, and work supervision has been proven to reduce workplace accidents and increase healthcare worker productivity. However, OHS implementation still faces challenges, such as limited resources, lack of employee awareness, and a suboptimal oversight system.In conclusion, the integrated implementation of OHS within human resource management in hospitals is a strategic step in creating a safe and healthy work environment. Efforts to improve the quality of human resources through OHS training, strengthening internal regulations, and fostering a culture of occupational safety are necessary to minimize occupational risks and improve the overall quality of healthcare services.
Keywords: OHS, Human Resource Management, Hospitals, Occupational Risks, Occupational Safety
Abstrak: Rumah sakit merupakan lingkungan kerja dengan tingkat risiko tinggi yang melibatkan paparan bahaya biologis, kimia, fisik, ergonomi, dan psikososial yang dapat berdampak pada keselamatan tenaga kerja. Oleh karena itu, implementasi Keselamatan dan Kesehatan Kerja (K3) dalam manajemen Sumber Daya Manusia (SDM) menjadi aspek penting dalam upaya pencegahan risiko kerja. Penelitian ini bertujuan untuk menganalisis implementasi K3 dalam manajemen SDM di rumah sakit sebagai strategi preventif terhadap kecelakaan kerja dan penyakit akibat kerja. Metode yang digunakan adalah studi literatur dengan pendekatan deskriptif kualitatif, melalui penelaahan berbagai artikel ilmiah nasional yang dipublikasikan dalam kurun waktu 2022–2025 dan tersedia dalam format PDF. Hasil kajian menunjukkan bahwa penerapan K3 yang efektif dipengaruhi oleh beberapa faktor utama, antara lain komitmen manajemen, pelatihan dan kompetensi SDM, ketersediaan sarana prasarana, serta budaya keselamatan kerja di lingkungan rumah sakit. Selain itu, integrasi K3 dalam fungsi manajemen SDM seperti rekrutmen, pelatihan, penilaian kinerja, dan pengawasan kerja terbukti mampu menurunkan angka kecelakaan kerja serta meningkatkan produktivitas tenaga kesehatan. Namun demikian, masih terdapat kendala dalam implementasi K3, seperti keterbatasan sumber daya, kurangnya kesadaran tenaga kerja, dan belum optimalnya sistem pengawasan.
Kesimpulannya, implementasi K3 yang terintegrasi dalam manajemen SDM di rumah sakit merupakan langkah strategis dalam menciptakan lingkungan kerja yang aman dan sehat. Upaya peningkatan kualitas SDM melalui pelatihan K3, penguatan regulasi internal, serta peningkatan budaya keselamatan kerja perlu terus dilakukan guna meminimalkan risiko kerja dan meningkatkan kualitas pelayanan kesehatan secara keseluruhan.
Kata kunci: K3, Manajemen SDM, Rumah Sakit, Risiko Kerja, Keselamatan Kerja
Abstract:Abstract: The growth of semi-structured and unstructured data on web pages has increased the need for efficient, autonomous, and structure-aware methods for structured data extraction. This study proposes a mobile agent…
approach based on the BDI (Belief-Desire-Intention) architecture as an alternative to conventional wrappers for structured web data extraction. The system was developed using the JADE (Java Agent Development Framework) platform and the Prometheus methodology, involving two main agents: AgenDownload, responsible for migration and retrieval of web page content, and AgenEkstraksi, which implements the Simple Tree Matching and MDR (Mining Data Regions) algorithms to identify and extract flat data records. Experiments were conducted on 40 web pages from Indonesian and English online bookstores. The results show an average recall of 73%, indicating the system capability to locate most target records, while the average precision of 15.59% indicates that further refinement is needed in post-extraction filtering. The mobile agent approach remains relevant in distributed network scenarios with bandwidth constraints, near-source processing requirements, and autonomous operation needs. This study contributes a multi-role agent system architecture that can be adapted for automated web data extraction.
Keywords: web data extraction; mobile agent; BDI architecture; Simple Tree Matching; JADE
Abstrak: Pertumbuhan data semi-terstruktur dan tidak terstruktur pada halaman web meningkatkan kebutuhan terhadap metode ekstraksi data terstruktur yang efisien, otonom, dan dapat beradaptasi terhadap variasi struktur HTML. Penelitian ini mengajukan pendekatan berbasis mobile agent dengan arsitektur BDI (Belief-Desire-Intention) sebagai alternatif terhadap wrapper konvensional dalam proses ekstraksi data terstruktur pada halaman web. Sistem dibangun menggunakan platform JADE (Java Agent Development Framework) dan metodologi Prometheus, melibatkan dua agen utama: AgenDownload yang bertugas melakukan migrasi dan pengambilan konten halaman web, serta AgenEkstraksi yang mengimplementasikan algoritma Simple Tree Matching dan MDR (Mining Data Regions) untuk mengidentifikasi serta mengekstrak flat data records. Pengujian dilakukan terhadap 40 halaman web toko buku daring berbahasa Indonesia dan Inggris. Hasil menunjukkan rata-rata recall sebesar 73%, yang mengindikasikan kemampuan sistem dalam menemukan sebagian besar target data, sedangkan rata-rata precision sebesar 15,59% menunjukkan perlunya penyempurnaan mekanisme pemfilteran hasil ekstraksi. Pendekatan mobile agent tetap relevan pada skenario jaringan terdistribusi dengan keterbatasan bandwidth, kebutuhan pemrosesan dekat sumber data, dan operasi otonom. Kontribusi penelitian ini adalah rancangan arsitektur sistem agen multi peran yang dapat diadaptasi untuk kebutuhan ekstraksi data web secara otomatis.
Kata Kunci: ekstraksi data web; mobile agent; arsitektur BDI; Simple Tree Matching; JADE
Abstract:Abstract: The development of disaster-related information in the digital space is very rapid and is often detected earlier through social media than through official government channels. This condition highlights the need…
d for a system capable of detecting disaster-related issues quickly and dynamically across various digital platforms. This study aims to develop a dynamic disaster issue detection model based on IndoBERTweet with a Temporal Attention Mechanism using multiplatform big text data in Indonesia. The research methodology includes collecting textual data from various digital platforms such as Twitter, YouTube, TikTok, Quora, and Medium. The data are processed through preprocessing stages using Natural Language Processing (NLP) techniques, timestamp extraction to obtain temporal information, and the application of fine-grained labeling for more detailed classification of disaster-related issues. Subsequently, the IndoBERTweet model is trained with a Temporal Attention Mechanism to capture the relationship between textual context and temporal dynamics in the development of disaster-related issues. The expected results of this research are a model capable of dynamically detecting disaster-related issues by considering informal language contexts and temporal changes. This model is expected to support early warning systems and data-driven disaster management decision-making in Indonesia.
Keywords: Disaster Issue Detection; Social Media Text Analysis; Multiplatform Big Data; IndoBERTweet; Temporal Attention.
Abstrak: Perkembangan informasi kebencanaan di ruang digital berlangsung sangat cepat dan sering kali lebih dahulu terdeteksi melalui media sosial dibandingkan melalui kanal resmi pemerintah. Kondisi ini menunjukkan perlunya sistem yang mampu mendeteksi isu bencana secara cepat dan dinamis dari berbagai platform digital. Penelitian ini bertujuan mengembangkan model deteksi isu bencana dinamis berbasis IndoBERTweet dengan Temporal Attention Mechanism pada big data teks multiplatform di Indonesia. Metode penelitian meliputi pengumpulan data teks dari berbagai platform digital seperti Twitter, YouTube, TikTok, Quora, dan Medium. Data diproses melalui tahapan preprocessing menggunakan teknik Natural Language Processing (NLP), ekstraksi timestamp untuk memperoleh informasi temporal, serta penerapan fine-grained labeling untuk klasifikasi isu bencana yang lebih rinci. Selanjutnya, model IndoBERTweet dilatih dengan Temporal Attention Mechanism untuk menangkap hubungan antara konteks teks dan dinamika waktu dalam perkembangan isu bencana. Hasil penelitian diharapkan menghasilkan model yang mampu mendeteksi isu bencana secara dinamis dengan mempertimbangkan konteks bahasa informal dan perubahan waktu. Model ini diharapkan mendukung sistem peringatan dini dan pengambilan kebijakan kebencanaan berbasis data di Indonesia.
Kata kunci: Deteksi Isu Bencana; Analisis Teks Media Sosial; Big Data Multiplatform; IndoBERTweet; Temporal Attention.
Abstract:Abstract: The phenomenon of brain rot poses a serious threat to the decline of students' cognitive function due to excessive exposure to low-quality digital content. This study aims to analyze the severity of brain rot and…
nd identify the most dominant digital behavioral factors of this phenomenon among students. As a solution to predict the level of risk quantitatively, this study implemented a machine learning approach using the Random Forest Regressor method. Data were collected from 500 student respondents through observation and questionnaires covering variables such as scrolling duration, app switching, GPA, study time, and cognitive symptoms. The test results showed that the model has not achieved optimal performance with an R2-Score of -0.177, RMSE 35.41, and MAE 31.1661. The low accuracy was influenced by inconsistencies in input data units and weak feature correlation in capturing non-linear patterns in the dataset. The study concluded that although scrolling duration was identified as the main influencing factor, the Random Forest model experienced high bias (underfitting). Therefore, hyperparameter optimization and data quality improvement are needed for future use.
Keyword: brainrot; students; machine learning; random forest regressor; digital behavior.
Abstrak: Fenomena brainrot (pembusukan otak) menjadi ancaman serius bagi penurunan fungsi kognitif mahasiswa akibat paparan konten digital yang berlebihan dan tidak berkualitas. Penelitian ini bertujuan untuk menganalisis tingkat keparahan brainrot serta mengidentifikasi faktor perilaku digital yang paling mendominasi fenomena tersebut pada kalangan mahasiswa. Sebagai solusi untuk memprediksi tingkat risiko secara kuantitatif, penelitian ini mengimplementasikan pendekatan machine learning dengan metode Random Forest Regressor. Data dikumpulkan dari 500 responden mahasiswa melalui observasi dan kuesioner yang mencakup variabel durasi scrolling, app switching, IPK, lama waktu belajar, dan gejala kognitif. Hasil pengujian menunjukkan bahwa model belum mencapai performa optimal dengan nilai R2-Score sebesar -0,177, RMSE 35,41, dan MAE 31,1661. Rendahnya akurasi dipengaruhi oleh ketidakkonsistenan satuan data input serta korelasi fitur yang kurang kuat dalam menangkap pola non-linear pada dataset. Simpulan penelitian menunjukkan bahwa meskipun durasi scrolling teridentifikasi sebagai faktor pengaruh utama, model Random Forest mengalami high bias (underfitting) sehingga diperlukan optimasi hyperparameter dan penyempurnaan kualitas data untuk penggunaan di masa mendatang..
Kata kunci: brainrot; mahasiswa; machine learning; random forest regressor; perilaku digital.