Abstract:This article aims to explore and understand how Generation Z in Medan responds to and interacts with music broadcasted by Radio Prambors, a popular radio station among young people. This study is crucial for identifying…
the dynamics between the musical preferences of the younger generation and the way radio content is presented in the digital era, as well as its implications for broadcasting strategies. Using a qualitative approach, this research involves in-depth interviews, observations, and document analysis to gather data from Generation Z listeners. Content analysis is used to interpret audience reactions to the music programming, with a source triangulation approach to enhance data validity, combining interviews, social media content analysis, and listener surveys. The research findings reveal three categories of audience reception: hegemony, where listeners fully accept the music choices presented; negotiation, where listeners are satisfied but offer suggestions for a more diverse genre presentation; and opposition, where some less featured music genres cause dissatisfaction among listeners. Furthermore, the interaction between listeners and the radio station shows that Radio Prambors Medan functions not only as a music provider but also as a mediator in a broader communication network that influences the musical tastes and preferences of its listeners. The discussion in this study integrates Manuel Castells' Network Society theory to explain how social and digital networks affect media consumption and interaction among Generation Z, highlighting the crucial role of digital technology adaptation in radio programming strategies. By understanding these dynamics, Radio Prambors Medan can more effectively craft strategies that appeal to its listeners, ultimately maintaining its relevance in the rapidly changing media landscape. The recommendations resulting from this study emphasize the importance of broader digital technology use and more intensive social media interaction to enhance listener engagement and enrich their listening experience.
Abstract:This study investigates the influence of artificial intelligence (AI) integration on strategic financial management in large corporations. Focusing on a sample of 20 Fortune 500 companies from diverse industries, the research…
earch employs a quantitative, descriptive-analytical approach utilizing secondary data from financial reports and AI system logs. The findings reveal that AI adoption significantly enhances forecasting accuracy, risk identification, and operational efficiency, while also enabling financial managers to redirect resources toward creative and strategic initiatives. However, the study also identifies challenges related to data quality, ethical considerations, and skill gaps in AI utilization. The results highlight the importance of a balanced approach that combines AI-driven insights with managerial intuition to maximize value creation in the digital age.
Abstract:Becoming a great entrepreneur in the global era is not as easy as flipping the palm of your hand. A successful and happy great entrepreneur is not solely determined by high academic performance, a high IQ, or attending a…
prestigious school. In fact, GPA, IQ, and attending a top school do not even rank among the top ten factors that guarantee someone will become a great and successful entrepreneur. What truly determines entrepreneurial greatness and success are soft skills. The top 10 soft skills referred to are: Honesty (Being honest with all people) ranked first, followed by Strong Discipline (Being well-disciplined); Sociability (Getting along with people); Support from a Partner (Having a supportive spouse); Hard Work (Working harder than most people); Passion for the Work (Loving my career/business); Leadership (Having strong leadership qualities); Competitive Personality (Having a very competitive spirit/personality); Well-Organized Life (Being very well-organized); and finally, Ability to Sell Ideas (Having an ability to sell my ideas/products). These ten soft skills will become even more meaningful when accompanied by a good heart, positive thinking, wise words, good deeds, good habits, positive culture, and strong character—all of which shape good fortune. Ultimately, good fortune begins with a good heart.
Abstract:The advanced innovation has essentially expanded the danger of cybercrime, driving to issues such as personality robbery, work misfortune, and disturbances to basic framework. In the interim, cybercriminals proceed to create…
eate modern procedures and methodologies in committing their criminal acts. The reason to depict the degree of cybercrime dangers in Indonesia. Common cyber dangers in Indonesia incorporate malware assaults, refusal of benefit (DoS), dispersed refusal of benefit (DDoS), and phishing. These dangers are frequently exacerbated by a need of mindfulness and instruction almost cybersecurity and constrained requirement of cybercrime laws. In conclusion, cybercrime dangers in Indonesia are right now considered genuine and profoundly unsafe due to their potential to make national issues. The action was carried out through the arrangement of socialization counseling with address and discourse strategies, as well as address and reply with members. Accommodation of fabric within the frame of control point slides that are displayed amid the action and conveyed to instructors as members . Therefore, it is crucial to recognize the importance of understanding these issues cybercrime anticipation socialization has to be passed on to the educator SD Negeri 132408 di Tanjungbalai Selatan. With this preparing, it is trusted that they can gotten to be web clients who can maintain a strategic distance from the potential to ended up culprits or casualties of cyber wrongdoings.
Keywords: cybercrime; socialization; prevention
Abstrak: Pengembangan teknologi digital telah secara signifikan meningkatkan risiko cybercrime karena mempengaruhi pencurian identitas, kehilangan tempat kerja, dan infrastruktur kritis. Sementara itu, pelaku cybercrime terus mengembangkan teknologi dan strategi baru untuk inspeksi perilaku kriminal. Ini dilakukan dengan memeriksa dan menganalisis literatur dalam kaitannya dengan dengan cybercrime di Indonesia. Temuan ini menunjukkan bahwa publikasi teknologi dari di Indonesia telah mengambil ancaman terbaru dan dampak pada berbagai aspek kehidupan dan bisnis. Penulis menyimpulkan bahwa cybercrime di Indonesia diklasifikasikan sebagai serius (sangat berbahaya) karena dapat menyebabkan masalah nasional. Kegiatan ini dilakukan dengan pemberian sosialisasi dengan menggunakan metode ceramah dan diskusi, serta bertanya jawab bersama peserta. Bentuk dalam penyampaian materi yaitu slide powerpoint yang dipaparkan dan di edarkan pada saat kegiatan berlangsung. Oleh karena itu, pemahaman mengenai sosialisasi pencegahan cybercrime sangat penting untuk disampaikan kepada para guru di SD Negeri 132408, Tanjungbalai Selatan. Melalui pelatihan ini, diharapkan mereka dapat menjadi pengguna internet yang lebih bijak dan mampu menghindari risiko sebagai pelaku maupun korban kejahatan cybercrime.
Kata kunci: cybercrime; sosialisasi; pencegahan
Abstract:Abstract: Manual meeting documentation risks losing important information due to cognitive fatigue. Although automated summarization models have evolved, integrated end-to-end systems for Indonesian spoken language remain…
n highly limited. This study aims to design and evaluate an end-to-end automated meeting summarization architecture that directly integrates Automatic Speech Recognition (ASR) via OpenAI Whisper for transcription and the IndoT5 language model for abstractive summarization. IndoT5 was fine-tuned using a dataset of 486 Indonesian spoken language transcript pairs. Testing was conducted on a CPU infrastructure using MP4, MP3, and WAV formats. Results show the optimal fine-tuning configuration significantly improved accuracy, achieving ROUGE-1 (0.4167), ROUGE-2 (0.1973), and ROUGE-L (0.2701) scores. Computationally, the system achieved a Real-Time Factor below 1, processing data faster than the actual recording duration. Conclusively, integrating Whisper and IndoT5 shows potential in producing coherent meeting summaries with lightweight computational overhead, making it viable for local infrastructure implementation to ensure data privacy.
Keywords: abstractive summarization; ASR; end-to-end pipeline; IndoT5; real-time factor
Abstrak: Dokumentasi rapat manual rentan menghilangkan informasi penting akibat keterbatasan kognitif. Meskipun model peringkas otomatis telah berkembang, implementasi sistem terintegrasi (end-to-end) khusus percakapan lisan berbahasa Indonesia masih sangat terbatas. Penelitian ini bertujuan merancang dan mengevaluasi arsitektur peringkas rapat otomatis end-to-end yang mengintegrasikan langsung Automatic Speech Recognition (ASR) melalui OpenAI Whisper untuk transkripsi dan model bahasa IndoT5 untuk peringkasan abstraktif. Adaptasi domain dilakukan melalui fine-tuning IndoT5 menggunakan 486 pasang dataset transkrip lisan berbahasa Indonesia. Pengujian pada infrastruktur CPU menggunakan format MP4, MP3, dan WAV. Hasil pengujian menunjukkan konfigurasi fine-tuning optimal berhasil meningkatkan akurasi, dengan skor ROUGE-1 (0,4167), ROUGE-2 (0,1973), dan ROUGE-L (0,2701). Sistem mendemonstrasikan efisiensi komputasi dengan nilai Real-Time Factor di bawah 1, mengindikasikan waktu pemrosesan lebih cepat dari durasi rekaman asli. Kesimpulannya, integrasi Whisper dan IndoT5 menunjukkan potensi dalam menghasilkan ringkasan yang koheren dengan beban komputasi ringan, sehingga layak diimplementasikan pada infrastruktur lokal organisasi untuk menjaga privasi data.
Kata kunci: ASR; end-to-end pipeline; IndoT5; peringkasan abstraktif; real-time factor
Abstract:Abstract: Accurate forecasts of tea harvest production are important for workforce planning, factory operations, and marketing decisions, yet conventional estimation in plantations often relies on field experience and can…
n be biased and less adaptive to changing conditions. This study aims to develop a Random Forest Regression model to predict tea harvest production at the Bah Butong tea plantation using historical operational and climate-related data. The dataset consists of 60 monthly records (2020–2024) with six predictor variables: rainfall (mm), number of rainy days, pest level, weed level, number of harvested trees and land area. Data were split into 80% training (48 samples) and 20% testing (12 samples). Model hyperparameters were optimized using RandomizedSearchCV with RepeatedKFold cross-validation (5 folds, 3 repeats). The tuned model achieved MSE of 668,980,524.45, RMSE of 25,864.66 kg, MAE of 19,838.69 kg, and MAPE of 7.59% on the test set. The results indicate that the model can provide practical production estimates, with errors averaging about 7–8% of the actual production. Feature importance analysis shows that the number of harvested tea bushes and cultivated area contribute most to predictions. Future work should extend the historical period and incorporate time-based features (seasonality/lag) for improved forecasting.
Keywords: hyperparameter tuning; production prediction; random forest; regression; tea harvest
Abstrak: Perkiraan akurat produksi panen teh sangat penting untuk perencanaan tenaga kerja, operasional pabrik, dan keputusan pemasaran, namun estimasi konvensional di perkebunan seringkali bergantung pada pengalaman lapangan dan dapat bias serta kurang adaptif terhadap perubahan kondisi. Studi ini bertujuan untuk mengembangkan model Regresi Random Forest untuk memprediksi produksi panen teh di perkebunan teh Bah Butong menggunakan data operasional dan data terkait iklim historis. Dataset terdiri dari 60 catatan bulanan (2020–2024) dengan enam variabel prediktor: curah hujan (mm), jumlah hari hujan, tingkat hama, tingkat gulma, jumlah pokok panen, dan luas lahan. Data dibagi menjadi 80% data pelatihan (48 sampel) dan 20% data pengujian (12 sampel). Parameter model dioptimalkan menggunakan RandomizedSearchCV dengan validasi silang RepeatedKFold (5 lipatan, 3 pengulangan). Model yang telah disempurnakan mencapai MSE sebesar 668.980.524,45, RMSE sebesar 25.864,66 kg, MAE sebesar 19.838,69 kg, dan MAPE sebesar 7,59% pada set data uji. Hasil tersebut menunjukkan bahwa model dapat memberikan estimasi produksi yang praktis, dengan kesalahan rata-rata sekitar 7–8% dari produksi aktual. Analisis kepentingan fitur menunjukkan bahwa jumlah semak teh yang dipanen dan luas lahan budidaya paling berkontribusi pada prediksi. Pekerjaan selanjutnya harus memperpanjang periode historis dan menggabungkan fitur berbasis waktu (musiman/lag) untuk peramalan yang lebih baik.
Kata kunci: panen teh; prediksi produksi; random forest; regresi; tuning parameter
Abstract:YouTube has become a major platform for public discourse in Indonesia, yet large-scale sentiment analysis of its comments remains challenging due to dynamic content, informal language, and limited labeled data. This study…
y proposes a Selenium–IndoBERT pipeline for sentiment analysis of Indonesian YouTube comments using a pseudo-labeling approach. Data were collected from ten YouTube videos discussing the One Piece flag phenomenon, yielding 10,842 comments after preprocessing. Selenium was employed to extract comments from dynamic pages, while IndoBERT was fine-tuned on a small manually labeled dataset and used to generate pseudo-labels for unlabeled data. Model performance was evaluated using probabilistic metrics, including Coverage, Expected Calibration Error (ECE), and Brier Score. At a confidence threshold of 0.75, 78.5% of comments received pseudo-labels, with an ECE of 0.095 and a Brier Score of 0.174. Manual validation showed substantial agreement with human annotations (Fleiss’ kappa = 0.72). The results indicate that the proposed pipeline enables scalable and reliable sentiment analysis with minimal manual annotation.
Abstract:Abstract: This study discusses the optimization of RAG for a FAQ system in the field of information technology product security certification at BSSN. Although LLM generate reliable responses, they often lack up-to-date…
and domain-specific knowledge, which can be addressed through the RAG approach. This research aims to optimize a domain-specific RAG system by improving embedding performance, enhancing prompt robustness, and increasing retrieval accuracy. The research methods consist of three stages. The first stage involves fine-tuning the bge-m3 embedding model and evaluating its performance using MRR, Recall, and AUC. The second stage applies prompt engineering techniques, namely the SRSM and Autodefense, to mitigate direct-injection and escape-character prompt injection attacks. The third stage evaluates the proposed RAG system using Precision, Recall, and F1-Score metrics against four baseline models. The results of research show that the fine-tuned embedding model achieves higher performance than the original model, with MRR@1 and Recall@1 values of 0.80 and an AUC@100 of 0.7023. In addition, the proposed prompt engineering techniques demonstrate robustness against prompt injection attacks, while the overall RAG system attains a perfect Precision, Recall, and F1-Score of 1.00. In conclusion, the proposed approach effectively enhances retrieval accuracy, embedding quality, and system security, resulting in a more reliable RAG-based FAQ system for information technology product security certification.
Keywords: embedding fine-tuning; large language model; prompt engineering; prompt injection mitigation; retrieval-augmented generation
Abstrak: Studi ini membahas optimasi RAG untuk sistem FAQ di bidang sertifikasi keamanan produk teknologi informasi di BSSN. Meskipun LLM menghasilkan respons yang andal, mereka seringkali kurang memiliki pengetahuan terkini dan spesifik domain, yang dapat diatasi melalui pendekatan RAG. Penelitian ini bertujuan untuk mengoptimalkan sistem RAG spesifik domain dengan meningkatkan kinerja embedding, meningkatkan ketahanan prompt dan meningkatkan akurasi pengambilan. Metode penelitian terdiri dari tiga tahap. Tahap pertama melibatkan fine-tuning model embedding bge-m3 dan mengevaluasi kinerjanya menggunakan Mean Reciprocal Rank (MRR), Recall, dan AUC. Tahap kedua menerapkan teknik rekayasa prompt, yaitu Self- SRSM dan Autodefense, untuk mengurangi serangan direct-injection dan escape-character prompt injection. Tahap ketiga mengevaluasi sistem RAG yang diusulkan menggunakan metrik Presisi, Recall, dan F1-Score terhadap empat model dasar. Hasil penelitian menunjukkan bahwa model embedding yang disempurnakan mencapai kinerja yang lebih tinggi daripada model asli, dengan nilai MRR@1 dan Recall@1 sebesar 0,80 dan AUC@100 sebesar 0,7023. Selain itu, teknik rekayasa prompt yang diusulkan menunjukkan ketahanan terhadap serangan injeksi prompt, sementara sistem RAG secara keseluruhan mencapai Presisi, Recall, dan F1-Score sempurna sebesar 1,00. Kesimpulannya, pendekatan yang diusulkan secara efektif meningkatkan akurasi pengambilan, kualitas embedding dan keamanan sistem, menghasilkan sistem FAQ berbasis RAG yang lebih andal untuk sertifikasi keamanan produk teknologi informasi.
Kata kunci: penyempurnaan embedding; model bahasa besar; rekayasa prompt; mitigasi injeksi prompt; retrieval-augmented generation
Abstract:Abstract: The rapid growth of the digital music industry requires accurate music genre classification systems to enhance user experience in streaming services. This study compares a domain-specific Long Short-Term Memory…
(LSTM) network with three Large Language Models (LLMs)—HuBERT, WavLM, and WAV2Vec 2.0—for Music Genre Classification (MGC). The LSTM model was trained using Mel-spectrograms transformed from the GTZAN dataset, while the LLMs were fine-tuned using a smaller set of raw audio samples due to computational constraints. All models were tested on datasets with identical genre labels to ensure a fair evaluation. Results show that the LSTM model achieved the highest accuracy of 97.10%, outperforming HuBERT (86.00%), WavLM (83.00%), and WAV2Vec 2.0 (80.00%). The LSTM demonstrated superior generalization and stability without overfitting, while the LLMs struggled to differentiate between genres with similar acoustic characteristics. These findings indicate that general-purpose pre-trained models, although powerful, are less effective in music-specific tasks due to domain mismatch. Therefore, incorporating music-specific features and architectures remains essential for achieving higher accuracy and reliability in automatic genre classification systems.
Keywords: audio large language models; comparative deep learning; music genre classification.
Abstrak: Pertumbuhan industri musik digital yang pesat menuntut sistem klasifikasi genre musik yang akurat untuk meningkatkan pengalaman pengguna dalam layanan streaming. Penelitian ini dilatarbelakangi oleh perkembangan pesat model pembelajaran mendalam, khususnya jaringan LSTM dan model bahasa berskala besar LLM seperti HuBERT, WavLM, dan WAV2Vec 2.0, yang telah menunjukkan kemampuan representasi audio yang kuat. Tujuan penelitian ini ini membandingkan jaringan Long Short-Term Memory (LSTM) khusus domain dengan tiga model Large Language Models (LLM)—HuBERT, WavLM, dan WAV2Vec 2.0—untuk tugas Klasifikasi Genre Musik (MGC). Metode penelitian melibatkan pelatihan LSTM menggunakan data Mel-spectrogram hasil transformasi dari dataset GTZAN, sementara LLM disesuaikan (fine-tuning) menggunakan data audio mentah dalam jumlah lebih kecil karena keterbatasan komputasi. Seluruh model diuji pada dataset dengan label genre yang sama untuk memastikan evaluasi yang adil. Hasil penelitian menunjukkan bahwa model LSTM mencapai akurasi tertinggi sebesar 97,10%, sedangkan model HuBERT, WavLM, dan WAV2Vec 2.0 masing-masing memperoleh 86,00%, 83,00%, dan 80,00%. Model LSTM menunjukkan kemampuan generalisasi yang lebih baik tanpa overfitting, sedangkan model LLM cenderung kesulitan membedakan genre dengan karakteristik akustik yang mirip. Kesimpulan penelitian ini adalah ketidaksesuaian domain secara signifikan membatasi performa model umum saat diterapkan pada tugas berbasis musik. Oleh karena itu, penggunaan fitur dan arsitektur khusus musik sangat penting dalam membangun sistem klasifikasi genre yang lebih akurat.
Kata kunci: klasifikasi genre musik; model bahasa besar; perbandingan pembelajaran mendalam.