Article Per Year (5 Year)
p-Index From 2021 - 2026
0.835
P-Index
This Author published in this journals
All Journal JOMLAI: Journal of Machine Learning and Artificial Intelligence
Victor Asido Elyakim P
STIKOM Tunas Bangsa Pematangsiantar
Computer Science & IT Engineering

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

 1 
Diagnosis of Skin Diseases Using Artificial Neural Networks with Backpropagation Algorithm Dony Jordan Pangomoan Sirait; Angga Priandi; Yemima Pepayosa Sembiring; Alyah Octafia; Victor Asido Elyakim P
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 4 No. 1 (2025): Maret 2025
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v4i1.5643

Abstract

Skin health is a vital aspect as it functions as the body's primary protector from the external environment. Various skin diseases can arise due to infections, allergies, autoimmune disorders, or environmental factors, and often exhibit similar symptoms, making diagnosis difficult. Artificial intelligence technology, such as Artificial Neural Networks (ANN), offers an innovative solution for accurate diagnosis. One popular ANN method is Backpropagation, which updates network weights iteratively based on the errors produced. This research focuses on applying the Backpropagation algorithm to diagnose skin diseases based on patient symptoms. With a binary data-based system and training using Backpropagation, this system is expected to accurately map symptoms to types of skin diseases. The methodology involves problem identification , data collection (types of skin diseases and symptoms, encoded in binary), dataset and diagnosis rule formation , ANN design (input, hidden, and output layers) , and training and testing using binary data and one-hot encoding. The results indicate that the application of ANN with Backpropagation is effective in assisting the automatic diagnosis process for skin disease cases , achieving an accuracy of 90%. This demonstrates the significant potential of this method in automated medical expert systems.
Analysis of the Impact of Balance Between Work and Study on Student Learning Productivity STIKOM Tunas Bangsa Pematangsiantar Hafizah Rahmi Lubis; Wanda Eka Nugraha; Zaskia Aulia Zahra; Ferdinand Saragih; Victor Asido Elyakim P
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 4 No. 1 (2025): Maret 2025
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v4i1.5644

Abstract

Working while studying has become a common phenomenon among students, especially due to the increasing cost of living and education. A balance between study and work is essential to maintain learning productivity, especially for students who are also active in campus organizations. However, many students have difficulty managing their time, which can lead to stress, fatigue, and decreased academic quality. This study highlights the challenges faced by working students, such as high workload, academic demands, and lack of rest time. With good time management, such as creating a clear schedule and utilizing free time effectively, students can achieve a better balance between study and work. Therefore, it is important for students to develop optimal time management strategies to maintain learning productivity and achieve academic success
Application of Backpropagation Algorithm for Prediction of Sales Results of Basic Foodstuffs at Artha Water Store Dwi Safitri Ramadhani; Abdul Ghani Ardiansyah; Damar Arya Prayoga; Riko ILham Nandika; Victor Asido Elyakim P
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 4 No. 1 (2025): Maret 2025
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v4i1.5764

Abstract

Companies need to implement sales growth forecasting strategies to create a balance between inventory and sales needs. Without this effort, an imbalance between inventory and sales can cause losses for the company, both in terms of finance and customer satisfaction. As a business entity engaged in the sale of basic necessities, Artha Water is committed to managing its business seriously in order to achieve profits and meet customer needs optimally. However, the development of consumer consumption patterns and sales growth lines at Artha Water which are fluctuating (up and down) make it quite difficult for the cooperative to balance inventory with demand for goods from consumers. By utilizing science in Artificial Neural Networks, we can predict future income using the Backpropagation Algorithm. From the previous description, the author concludes that from the results of the study with the best architecture experiments, namely 12-10-1 to predict sales growth at the Artha Water Store in 2024, it shows an accuracy result of 92%, MSE training of 0.06031588, that there is a significant difference, in other words, sales growth at the Artha Water Store will increase in 2024. With a total sales result of basic necessities at the Artha Water Store for 2024 of IDR 336,930,000.
Analysis of Egg Production Forecasting by Province in Indonesia Using the ARIMA Algorithm Khaswa Giovani Simanungkalit; Muhammad Fikri Azhari; Muhammad ihsan Raditya; Indra Lesmana Putra; Victor Asido Elyakim P
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 4 No. 1 (2025): Maret 2025
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v4i1.5765

Abstract

The production of chicken eggs in various regions of Indonesia shows significant variations over time, making it necessary to apply an appropriate predictive approach to support national food planning and distribution strategies. This study employs the ARIMA (AutoRegressive Integrated Moving Average) method to forecast regional chicken egg production based on secondary data from 2018 to 2024. The research steps include data collection, stationarity testing, model parameter determination, as well as the modeling process and result evaluation. The predictions indicate that total national chicken egg production will experience a significant increase, from 12.5 billion eggs in 2025 to 18.57 billion eggs in 2026. Provinces on the island of Java, such as East Java, Central Java, and West Java, are expected to remain the main production centers. Meanwhile, provinces in eastern Indonesia show less stable prediction results, indicating the need for improved data quality and the application of more adaptive models. Overall, the ARIMA model is considered effective for modeling short-term trends, although it has limitations in handling data with high fluctuations.
Diagnosis of Gastric Disease Based on Artificial Neural Network with Hebb Rule Algorithm Victor Asido Elyakim P; Alyah Octafia; Yemima Pepayosa Sembiring; Dony Jordan Pangomoan Sirait; Angga Priandi
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 4 No. 3 (2025): September 2025
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v4i3.6543

Abstract

Gastric disorders are among the most common health problems faced by society, often caused by irregular eating habits, unhealthy lifestyles, and high stress levels. The symptoms are diverse, ranging from abdominal pain and nausea to weight loss, making accurate and timely diagnosis essential to prevent more serious complications. This study aims to develop a diagnostic system for gastric diseases using Artificial Neural Networks (ANN) with the Hebb Rule algorithm, a learning principle that strengthens the connections between neurons when they are activated simultaneously. The research utilized binary-encoded data consisting of ten types of gastric diseases and twenty associated symptoms to establish patterns of correlation between symptoms and diagnoses. The results demonstrate that the system successfully recognized all test data with outcomes consistent with the expected targets, proving that the Hebb Rule is effective in mapping symptom-disease relationships even when applied to simple binary data. These findings highlight the practicality and efficiency of the Hebb Rule in building an intelligent diagnostic framework, while also showing its potential for further development with more complex datasets, such as symptom severity levels or laboratory test results. Ultimately, this research contributes to the advancement of smart medical systems that can support both healthcare professionals and the general public in performing early detection of gastric diseases quickly, accurately, and effectively.
Co-Authors Abdul Ghani Ardiansyah Alyah Octafia Angga Priandi Damar Arya Prayoga Dony Jordan Pangomoan Sirait Dwi Safitri Ramadhani Ferdinand Saragih Hafizah Rahmi Lubis Indra Lesmana Putra Khaswa Giovani Simanungkalit Muhammad Fikri Azhari Muhammad ihsan Raditya Riko ILham Nandika Wanda Eka Nugraha Yemima Pepayosa Sembiring Zaskia Aulia Zahra