Claim Missing Document
Check
Articles

Found 4 Documents
Search

Identifikasi Chatbot dalam Meningkatkan Pelayanan Online Menggunakan Metode Natural Language Processing Muliyono
Jurnal Informatika Ekonomi Bisnis Vol. 3, No. 4 (December 2021)
Publisher : SAFE-Network

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (694.814 KB) | DOI: 10.37034/infeb.v3i4.102

Abstract

Chatbot is a software with artificial intelligence that can imitate human conversations through text messages or voice messages. This chatbot can convey information, according to the knowledge that has been given previously. Helping the limitations of the academic section in answering questions posed by students. The method in this study was sourced from a questionnaire distributed to students at the Muhammadiyah University of West Sumatra. Based on the analysis of the questionnaire, there are 40 questions that are often asked by students to the academic section. Then it is processed using Natural Language Processing (NLP). Natural Language Processing is a branch of science from artificial intelligence that is able to study communication between humans and computers through natural language. The processing stage is to identify the intent, process the input and display the results according to the input. The results of the test using a questionnaire addressed to 227 students got a score of 3,55 with a very good predicate. Then do the test using 40 question and answer data. So, obtained 37 appropriate answers and 3 answers that are not in accordance with the percentage of answer accuracy generated from the chatbot is 92.5 percent. The results of this test have been able to respond to the questions asked by students. This chatbot can make it easier for students to get information with a very good level of accuracy
Pelatihan Pembuatan Konten Digital Percepat Digitalisasi Desa Cerdas bagi Masyarakat Nagari Singkarak Saputra, Rahmad; Putra, Yendi; Muliyono
ORAHUA : Jurnal Pengabdian Kepada Masyarakat Vol. 2 No. 02 (2025): Januari
Publisher : Faatuatua Media Karya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70404/orahua.v2i02.122

Abstract

Dalam era transformasi digital, penguasaan teknologi informasi dan kemampuan menciptakan konten digital yang relevan menjadi kebutuhan mendesak bagi masyarakat desa. Pelatihan bertajuk "Pelatihan Pembuatan Konten Digital Percepat Digitalisasi Desa Cerdas bagi Masyarakat Nagari Singkarak" ini diselenggarakan dengan tujuan untuk mempercepat proses digitalisasi desa cerdas di Nagari Singkarak melalui pelatihan pembuatan konten digital. Kegiatan ini melibatkan pelatihan langsung bagi masyarakat lokal, termasuk pemuda, perangkat desa, dan pelaku UMKM, dalam berbagai aspek pembuatan konten seperti pengambilan gambar, pengeditan video, hingga strategi pemasaran digital. Selain itu, pelatihan ini juga menghadirkan narasumber yang berpengalaman di bidang digital marketing dan konten digital, yang akan berbagi tips dan pengalaman praktis dalam meanfaatkan platform digital untuk mencapai target audiens yang lebih luas. Dengan pendekatan partisipatif, masyarakat dilatih untuk memanfaatkan media digital sebagai alat promosi potensi desa, seperti pariwisata, produk lokal, dan budaya tradisional. Hasil pelatihan menunjukkan peningkatan kemampuan peserta dalam menghasilkan konten yang kreatif dan efektif, serta meningkatkan kesadaran akan pentingnya digitalisasi dalam pembangunan desa. Pelatihan ini diharapkan dapat mendorong kemandirian masyarakat Nagari Singkarak dalam memanfaatkan teknologi untuk meningkatkan kesejahteraan dan daya saing desa di era digital.
Prediksi Arah Harga Cryptocurrency Menggunakan Hybrid Lstm Encoder dan Xgboost Head: Implementasi dan Evaluasi pada 10 Aset Digital Utama Ade Kurniawan; Muliyono; Hari Suriadi
Jurnal Ilmu Sosial, Ekonomi dan Pendidikan Vol. 1 No. 2 (2025): Oktober 2025
Publisher : Suria Academic Press

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The primary challenge in cryptocurrency price prediction lies in the market’s highly volatile, non-linear, and near–random walk behavior, which makes traditional predictive models unable to achieve consistent accuracy. This study aims to develop and evaluate a hybrid model combining Long Short-Term Memory (LSTM) and XGBoost to predict price direction and returns for ten major cryptocurrencies using daily data from 2023 to 2025. Historical data were processed through feature engineering, normalization, and sliding-window sequence construction, and the models were evaluated using TimeSeriesSplit to prevent data leakage. The results show that the hybrid model consistently outperformed both LSTM and XGBoost, achieving an average directional accuracy of 58.6%, significantly higher than the baselines (51.7% for LSTM and 53.6% for XGBoost). The average RMSE of 0.0289 indicates stable return predictions without systematic bias. Statistical validation through paired t-tests and McNemar tests confirmed the significance of the improvement at p < 0.001. A trading simulation using a 1-day holding period produced an annualized return of 41.5% with a Sharpe ratio of 1.12, outperforming the buy-and-hold strategy. These findings highlight that integrating LSTM’s temporal representation with XGBoost’s non-linear learning capabilities is an effective and computationally efficient approach for cryptocurrency price forecasting, offering practical value for the development of algorithmic trading systems.
Correlation of CBR Values and Mackintosh Probe on Clay Soil with Variations of Bentonite, Kaolin and Sand Nugroho, Soewignjo Agus; Fatnanta, Ferry; Wibisono, Gunawan; Muliyono
Journal of Geoscience, Engineering, Environment, and Technology Vol. 10 No. 4 (2025): JGEET Vol 10 No 04 : December (2025)
Publisher : UIR PRESS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25299/jgeet.2025.10.4.24252

Abstract

Coastal areas are typically characterized by non-uniform soil properties, often featuring soft, water-saturated soils with high plasticity, which frequently results in low soil bearing capacity. This study aims to investigate the relationship between California Bearing Ratio (CBR) values and Mackintosh Probe (MP) test results by utilizing a mixture of clay soil comprising bentonite and kaolin with sand in various compositions. These mixtures were prepared as laboratory test samples to simulate the soil conditions in these areas. The primary objective of this research is to develop a faster and more efficient alternative method for estimating soil bearing capacity in coastal regions. A total of 81 samples were prepared with variations in moisture content, compaction levels, and the composition of sand and clay mixtures. Testing was conducted using both CBR and MP methods. The analysis revealed a significant positive correlation between MP and CBR values, represented by the linear regression model: CBR = 0.7498 * MP, with a coefficient of determination (R²) of 0.9542. This indicates that approximately 95.42% of the variation in CBR values can be predicted from the MP test results. The model's accuracy was further validated through training and testing using 5 randomly selected data points from the sample set. The findings suggest that the Mackintosh Probe can serve as a preliminary tool for estimating soil bearing capacity in coastal areas, particularly in field conditions where laboratory equipment is limited. However, for broader applicability, further validation of this model is necessary to accommodate more complex soil conditions in the field.