Article Per Year (5 Year)
p-Index From 2021 - 2026
1.037
P-Index
This Author published in this journals
All Journal Journal of Applied Data Sciences JOMLAI: Journal of Machine Learning and Artificial Intelligence Kesatria : Jurnal Penerapan Sistem Informasi (Komputer dan Manajemen)
Dedy Hartama
STIKOM Tunas Bangsa, Pematangsiantar, Indonesia
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Engineering

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

 1 
Backpropagation Model in Predicting the Location of Prospective Freshman Schools for Promotion Optimization Muhammad Fahrur Rozi; Dedy Hartama; Ika Purnama Sari; Rafiqa Dewi; Zulia Almaida Siregar
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 1 (2022): March
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (791.3 KB) | DOI: 10.55123/jomlai.v1i1.161

Abstract

In carrying out promotions, it is also necessary to pay for the manufacture of brochures, banners and other promotional media to provide information to prospective students and attract prospective students to register. Determining the location of the promotion is one of the success factors in promotional activities. In this study, the Artificial Neural Network will be used to predict the location of the promotion. Backpropagation is one of the best artificial neural network methods used for prediction, this method is widely used by researchers in predicting a problem. The data analysis tool used is Matlab or what we call the (Matrix Laboratory) which is a program to analyze and compute numerical data, and Matlab is also an advanced mathematical programming language, which was formed on the premise of using the properties and forms of matrices. From the results of the algorithm used, it is expected to get good accuracy results with some architectural experiments later. So that this research can be an indicator to optimize promotions in the following year in order to attract prospective students to register for AMIK and STIKOM Tunas Bangsa Pematangsiantar
Determining Product Suitability using Rule-Based Model with C4.5 Algorithm Chintya Carolina Situmorang; Dedy Hartama; Irfan Sudahri Damanik; Jaya Tata Hardinata
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 2 No. 1 (2023): March
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v2i1.1923

Abstract

A hotel warehouse must have orderly, good, safe, comfortable, and usable procurement of goods. The common issue that occurs in a warehouse is damaged and unusable goods. The fluctuating production demand for goods sometimes leads to neglecting the quality of the goods in the warehouse. To determine usable goods, appropriate recommendations are needed. The C4.5 algorithm with data mining techniques is an appropriate recommendation for analyzing a large amount of data for classification. The data used in this study is the inventory data of Hotel Sapadia Pematangsiantar's warehouse. Implementing the C4.5 algorithm that produces a Decision Tree can assist the warehouse in determining which goods are still usable for hotel activities. This study resulted in the best variable from the rule model used to determine the feasibility of goods being the physical condition of the goods. The accuracy of the rule model generated from the C4.5 Algorithm modeling is 99.02% against the feasibility of goods.
Sistem Pendukung Keputusan Menentukan Produk Terbaik Menggunakan Metode Simple Additive Weighting Pada CV. Siantar Simalungun Sukse Evani Sitohang; Dedy Hartama; Agus Perdana Windarto
Kesatria : Jurnal Penerapan Sistem Informasi (Komputer dan Manajemen) Vol 2, No 3 (2021): Edisi Juli
Publisher : LPPM STIKOM Tunas Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/kesatria.v2i3.76

Abstract

CV. Siantar Simalungun Sukses is a form of business that sells various kinds of milk to meet the needs of the community, especially for the people of Pematangsiantar. But CV. Siantar Simalungun Sukses is not the only form of business in Pematangsiantar city, but there are still many supermarkets operating in the same field, so that this will lead to business competition between existing forms of business. To face the business competition, there are many ways that can be done by CV. Siantar Simalungun Success to maintain the quality of the milk products sold, including by analyzing the quality of these products, and price competitiveness.
Multiclass Skin Lesion Classification Algorithm using Attention-Based Vision Transformer with Metadata Fusion Mhd. Furqan; Norliza Katuk; Dedy Hartama
Journal of Applied Data Sciences Vol 7, No 1: January 2026
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v7i1.1017

Abstract

Early and accurate classification of skin lesions is essential for timely diagnosis and treatment of skin cancer. This study presents a novel multiclass classification framework that integrates dermoscopic images with clinical metadata using an attention-based Vision Transformer (ViT) architecture. The proposed model incorporates a mutual-attention fusion mechanism to jointly learn from visual and tabular inputs, augmented by a class-aware metadata encoder and imbalance-sensitive loss function. Training was conducted using the HAM10000 dataset over 30 epochs with a batch size of 32, utilizing the Adam optimizer and a learning rate of 0.0001. The model demonstrated superior performance compared to a ViT Baseline, achieving 93.4% accuracy, 92.2% F1-score, 0.95 AUC, and significant reductions in MAE and RMSE. Additionally, Grad-CAM visualizations confirmed the model’s ability to focus on diagnostically relevant regions, enhancing interpretability. These findings suggest that the integration of structured clinical information with transformer-based visual analysis can significantly improve classification robustness, particularly in underrepresented lesion types. However, the model’s current performance is evaluated only on the HAM10000 dataset, and its generalizability to other clinical or non-dermoscopic image sources remains to be validated. Future studies should therefore explore multi-institutional datasets and real-world deployment scenarios to assess robustness and scalability. The proposed framework offers a practical, interpretable solution for AI-assisted skin lesion diagnosis and demonstrates strong potential for clinical deployment.
Co-Authors Agus Perdana Windarto Chintya Carolina Situmorang Dewi, Rafiqa Evani Sitohang Ika Purnama Sari Irfan Sudahri Damanik Jaya Tata Hardinata Mhd. Furqan Muhammad Fahrur Rozi Norliza Katuk Zulia Almaida Siregar