cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Fifi Syafrina
Contact Email
jcbd@delitekno.co.id
Phone
+6287869230953
Journal Mail Official
jcbd@delitekno.co.id
Editorial Address
Jl. Lapangan Bola Gg Rosela No.3, Dalu XB, Tanjung Morawa, Kab. Deli Serdang, Prov. Sumatera Utara, Indonesia 20362
Location
Kab. deli serdang,
Sumatera utara
INDONESIA
Journal of Computers and Digital Business
Published by PT Delitekno Media Mandiri
ISSN : -     EISSN : 28303121     DOI : 10.56427
Core Subject : Science,
Computer Science & IT Decision Sciences, Operations Research & Management
Journal of Computers and Digital Business is an interdisciplinary and open access journal covering Computers and Digital Business. The Journal of Computers and Digital Business is open to submission from experts and scholars in the wide areas of Information System, Security, Artificial Intelligent , Cloud Computing, Machine Learning, Digital Business Technology and other areas listed in the focus and scope of this journal. Focus and Scope Information System Information Security Information Retrieval Geographic Information System Fuzzy Logics Genetic Algorithms Neural Networks Machine Learning Decision Support System Data Mining Cloud Computing E-Learning E-Goverment E-Commerce E-Business Digital Business Management Digital Business Technology Digital Business Analysis & Design Big Data & Business Intelligence Cyber Security for Digital Business
Arjuna Subject : Ilmu Komputer (semua kategori) - Aplikasi Ilmu Komputer
Articles 83 Documents
 5   6   7   8   9 
Perbandingan Support Vector Regression dan Long Short-Term Memory untuk Prediksi Harga Crude Palm Oil Berjangka Khoirun Nisa Harahap; Ratu Mutiara Siregar; Raden Aris Sugianto
Journal of Computers and Digital Business Vol. 5 No. 3 (2026): Articles in Press
Publisher : PT. Delitekno Media Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56427/jcbd.v5i3.996

Abstract

Indonesia sebagai produsen Crude Palm Oil (CPO) terbesar dunia memerlukan sistem prediksi harga yang akurat untuk mendukung pengambilan keputusan strategis. Penelitian ini membandingkan kinerja algoritma Support Vector Regression (SVR) dan Long Short-Term Memory (LSTM) dalam memprediksi harga CPO berjangka Bursa Malaysia menggunakan 1.426 data harian dari Investing.com periode 2 Januari 2020 hingga 30 Desember 2025. Data diproses melalui pembersihan, normalisasi Min-Max, dan pembentukan sekuens dengan lookback 30, lalu dibagi secara kronologis dengan rasio 80:20. SVR menggunakan kernel Radial Basis Function (RBF) dengan penyetelan parameter melalui GridSearchCV, sedangkan LSTM dibangun dengan arsitektur dua lapis dan Early Stopping. Evaluasi dengan RMSE dan MAPE menunjukkan kedua model mencapai akurasi sangat baik dengan MAPE di bawah 2% pada data uji. SVR memperoleh MAPE 1,29% dan RMSE 76,17 MYR/ton, sedangkan LSTM memperoleh MAPE 1,32% dan RMSE 75,22 MYR/ton. Perbedaan antar model sangat kecil sehingga tidak dapat diklaim signifikan secara statistik tanpa uji formal. Temuan ini mengindikasikan bahwa SVR dan LSTM memiliki kapabilitas setara untuk prediksi harga CPO jangka pendek berbasis data historis, dengan keterbatasan yang sama dalam menghadapi guncangan eksternal yang tidak terepresentasi pada variabel input.
Association Rule Mining with FP-Growth for Cross-Selling Recommendation in a Culinary MSME Bagas Ramadhan; Rudolf Januar; Firman Noor Hasan
Journal of Computers and Digital Business Vol. 5 No. 3 (2026): Articles in Press
Publisher : PT. Delitekno Media Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56427/jcbd.v5i3.1021

Abstract

The rapid growth of digitalization in Micro, Small, and Medium Enterprises (MSMEs) has led to a significant increase in daily sales transaction data. However, most culinary MSMEs utilize this data merely as transaction records rather than processing it into valuable insights for business decision-making. Kedai Mie Rahma faces challenges in determining promotional strategies, designing menu packages, and managing raw material inventories due to the lack of systematic analysis of customer purchasing patterns. This study aims to extract valid association rules from sales data using the FP-Growth algorithm to provide data-driven cross-selling recommendations. The methodology follows the Knowledge Discovery in Databases (KDD) framework, analyzing 2,500 historical transactions from the Point of Sales (POS) system. The results successfully identified significant purchasing patterns, with the strongest valid rule being the association between Sweet Snacks and Ngemie (Support: 6.4%, Confidence: 79.9%, Lift Ratio: 1.100). The findings transition the use of FP-Growth from a routine application into a practical decision-support tool, providing the MSME management with a strategic basis for menu bundling, cross-selling opportunities, and targeted inventory planning.
Optimization of LPG Distribution Using the Saving Matrix and Nearest Neighbor Methods Fairuz Naila Qosammah; Ismail Husein
Journal of Computers and Digital Business Vol. 5 No. 2 (2026)
Publisher : PT. Delitekno Media Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56427/jcbd.v5i2.1010

Abstract

The distribution of 3 kg LPG cylinders at UD. Tri Surya Efendi is currently conducted conventionally, without structured route planning, leading to fuel wastage, longer travel distances, and high operational costs. This research aims to determine the optimal distribution route by applying the Saving Matrix and Nearest Neighbor methods for 26 regular customers in the Helvetia–Marelan area with a total daily demand of 100 cylinders. Prior to optimization, customers were segmented into three geographic zones, and the distances between the depot and customer locations were obtained using Google Maps. The Saving Matrix method produced three routes spanning 28.06 km at a daily cost of Rp14,300, reducing operational costs by 64.25% compared to the initial Rp40,000. The Nearest Neighbor method produced four routes spanning 30.26 km at Rp15,130, reducing costs by 62.18%. The results show that the Saving Matrix method is more efficient in minimizing travel distance and distribution costs. This study contributes an empirical comparison of both methods in the context of small-scale LPG distribution with prior geographic zone segmentation, a procedure rarely documented in existing literature, and offers a practical reference for improving local distribution efficiency.

Page 9 of 9 | Total Record : 83

 5   6   7   8   9 

Filter by Year

2022 2026


Reset
Filter By Issues
All Issue Vol. 5 No. 3 (2026): Articles in Press Vol. 5 No. 2 (2026) Vol. 5 No. 1 (2026) Vol. 4 No. 3 (2025) Vol. 4 No. 2 (2025) Vol. 4 No. 1 (2025) Vol. 3 No. 3 (2024) Vol. 3 No. 2 (2024) Vol. 3 No. 1 (2024) Vol. 2 No. 3 (2023) Vol. 2 No. 2 (2023) Vol. 2 No. 1 (2023) Vol. 1 No. 2 (2022) Vol. 1 No. 1 (2022) More Issue