Article Per Year (5 Year)
p-Index From 2021 - 2026
2.56
P-Index
This Author published in this journals
All Journal Jurnal Sains dan Teknologi Jurnal Pengabdian Masyarakat Bidang Sains dan Teknologi JOMLAI: Journal of Machine Learning and Artificial Intelligence Jurnal Bersama Pengabdian Kepada Masyarakat
Victor Asido Elyakim P
STIKOM Tunas Bangsa
Religion Aerospace Engineering Agriculture, Biological Sciences & Forestry Arts Humanities Astronomy Automotive Engineering Biochemistry, Genetics & Molecular Biology Chemical Engineering, Chemistry & Bioengineering Chemistry Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Dentistry Earth & Planetary Sciences Economics, Econometrics & Finance Education Electrical & Electronics Engineering Energy Engineering Environmental Science Health Professions Immunology & microbiology Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Library & Information Science Materials Science & Nanotechnology Mathematics Mechanical Engineering Medicine & Pharmacology Neuroscience Nursing Physics Public Health Social Sciences Transportation Veterinary Other

Published : 19 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 19 Documents
Search

 1   2 
Public Sentiment Analysis of the Agrarian Conflict between PT TPL and the Toba Simalungun Indigenous Community Using the SVM Method Dian Yusri Andira; Deswita Maharani Harahap; Vibiola Br Damanik; Indah Frian Sari; Victor Asido Elyakim P
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 5 No. 2 (2026): Juni 2026
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v5i2.6605

Abstract

The agrarian conflict between PT Toba Pulp Lestari and the Toba Simalungun indigenous community has generated diverse public opinions on social media. This study aims to analyze public sentiment regarding the conflict using the Support Vector Machine (SVM) method based on TikTok comment data. A total of 1,751 comments were collected via the TikTok API and processed through cleaning, normalization, stopword removal, and stemming. Sentiment labeling was performed automatically with a lexical-based approach, followed by feature weighting using Term Frequency-Inverse Document Frequency (TF-IDF). The SVM model was used to classify public sentiment into two main categories, namely positive and negative. The results of the testing showed that the SVM model was able to achieve an accuracy of 80%, with excellent performance in detecting negative sentiment. Additional analysis through wordcloud visualization shows the dominant words in each sentiment category, which reinforces the model's classification results. The findingsof this study provide an objective picture of public opinion patterns on social media, while also demonstrating the potential application of machine learning-based sentiment analysis methods to understand public perceptions of other social issues in the future.
Implementation of SVM in Predicting Obesity Risk Based on Lifestyle and Dietary Patterns Adinda Febiola; Fahriya Ardiningrum; Michael Orlando A. Purba; Fernando Siahaan; 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.v4i1.5766

Abstract

Obesity is one of the global health issues that has seen a significant increase in recent decades. This condition is closely related to an unbalanced modern lifestyle, such as lack of physical activity, unhealthy eating patterns, and habits of smoking and alcohol consumption. This study aims to analyze the relationship between lifestyle and obesity risk, as well as to evaluate the effectiveness of the Support Vector Machine (SVM) method in predicting the level of obesity risk. The dataset used was obtained from the Kaggle platform, covering various variables such as age, gender, body mass index (BMI), eating habits, sleep patterns, and physical activity. Preprocessing was carried out through data normalization and encoding of categorical variables to ensure data readiness before being input into the model. The SVM model was trained using various training and testing data split ratios and showed a very high accuracy rate, even reaching 100% in some scenarios. These results demonstrate that SVM can effectively identify patterns in lifestyle data that contribute to obesity. Thus, the application of SVM can be a useful predictive tool for healthcare professionals in designing more accurate and efficient data-driven obesity prevention strategies.
Analysis of Airline Passenger Satisfaction Using the Rough Set Method Alisa Putri Amanda Nasution; Auralia Izmi; Aprillya Zahra Iswandy Lubis; Haya Atiqah Tampubolon; Victor Asido Elyakim P
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.v4i3.5946

Abstract

This study analyzes airline passenger satisfaction using the Rough Set method, an effective approach in handling complex data without requiring additional information such as probability. The main factors influencing customer satisfaction are identified based on data collected through questionnaires and analyzed using the attribute reduction method. The results show that flight punctuality, cabin crew service quality, and flight class type have a significant influence on customer satisfaction. From the survey results, 72% of respondents stated that they were satisfied, 18% were quite satisfied, and 10% were dissatisfied, with dissatisfaction generally related to flight delays and lack of comfortable facilities. The application of the Rough Set method has been proven to be able to identify passenger satisfaction patterns more accurately, so that it can be used by airlines to improve their service strategies.
Prediction of Poverty Levels in Indonesia Using the Tsukamoto Fuzzy Logic Method Aklima Laduna Ramadya; Tiara Dwi Lestari Purba; Ega Wahyu Andani; Baginda Faustine Sinaga; 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.v4i1.5955

Abstract

Poverty remains a fundamental issue and a primary focus in Indonesia's development. Conventional analysis often fails to provide an accurate picture due to the complexity of its underlying factors. This study aims to build a prediction model for poverty levels in Indonesia using the Tsukamoto fuzzy logic method, based on macroeconomic data from the Central Statistics Agency (BPS) for the years 2022 to 2024. Input variables include inflation rates, unemployment, and economic growth, with the output being the predicted poverty level in percentage. The fuzzy inference process involves fuzzification, rule base formation, fuzzy logic inference, and defuzzification. Data on the percentage of the poor population from BPS shows a decrease from 9.57% in 2022 to 9.27% in 2024. However, significant regional disparities and economic vulnerabilities persist due to global factors like inflation. Fuzzy logic, especially the Tsukamoto fuzzy method, is an adaptive approach capable of handling uncertainty and linguistic variables, while producing numerical outputs. The research results indicate that the fuzzy Tsukamoto model successfully predicts poverty levels with high accuracy, showing an average difference of less than 0.1% from actual data. This finding suggests that the Tsukamoto fuzzy method can be an effective predictive alternative in addressing socio-economic data uncertainties and supporting the formulation of more targeted policies.
Analysis of Unemployment Rate in Indonesia Using Fuzzy Inference System Tiara Dwi Lestari Purba; Aklima Laduna Ramadya; Ega Wahyu Andani; Baginda Faustine Sinaga; 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.v4i1.5956

Abstract

Unemployment is a complex problem that demands an analytical approach capable of handling data uncertainty. This study utilizes a fuzzy inference system to analyze unemployment rates in Indonesia, based on Central Statistics Agency (BPS) data for the 2023-2025 period. The fuzzy logic method was chosen due to its ability to handle linguistic variables and uncertainty in classifying unemployment levels. Input variables include education level, age group, and geographical area, while the output is a classification of unemployment risk (low, medium, high). The fuzzy inference process involves fuzzification, rule base formation, fuzzy logic inference, and defuzzification. BPS data indicates that the Open Unemployment Rate (TPT) experienced a consistent downward trend from 5.45% in February 2023 to 4.76% in February 2025. Nevertheless, the complexity of unemployment requires a flexible approach that can capture nuances of uncertainty, which conventional methods are unable to address. The research results show that the fuzzy inference system is capable of classifying unemployment levels with an accuracy of 87.3%. The highest unemployment rate is found in the 15-24 age group and among high school/vocational school graduates. This system can serve as a decision-making tool for the government in formulating more targeted employment policies.
Analysis of Open Unemployment Rate Prediction Using Backpropagation Method P.A.M. Zidane R.W.P.P. Zer; Dimas Prayogi; M Arif Y Sinaga; Olivia Diwani Saragih; Victor Asido Elyakim P
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 4 No. 2 (2025): Juni 2025
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v4i2.5957

Abstract

The open unemployment rate (TPT) is one of the important indicators in assessing the economic health of a region. This study aims to develop an accurate prediction model for the open unemployment rate using the backpropagation algorithm, as well as to evaluate the factors that influence the prediction. Accurate TPT prediction can help the government and policy makers in designing strategies to alleviate unemployment based on the results of the analysis of the developed model. This study aims to analyze and predict the Open Unemployment Rate (TPT) in various provinces in Indonesia in 2024 to 2026 using an artificial neural network model with the Backpropagation algorithm. Based on the test results, the 3-6-1 architecture model showed a prediction ability with 100% accuracy, while other architectures also gave very good results, with 100% accuracy for the 3-3-1 model and 97.06% for the 3-12-1 model. The TPT prediction results show that the unemployment rate is predicted to continue to increase from year to year, indicating the potential for an increase in the number of unemployed in the future. On the other hand, the accuracy analysis shows that each architecture produces different results, with the 3-6-1 architecture producing a longer time for the testing process, but still providing optimal accuracy. This finding illustrates that choosing the right architecture greatly affects the accuracy and efficiency in predicting TPT, which can be an important basis in formulating policies to eradicate unemployment in Indonesia.
Trend of Increasing Health Budget in Indonesia's State Budget During COVID-19 Pandemic Victor Asido Elyakim P; Bagas Adi Nata; M.Alfathan Haris; Mirza Afif Pradivta; Muhammad Rizky Ramadhan
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 4 No. 2 (2025): Juni 2025
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v4i2.5958

Abstract

This research aims to analyze trends and allocation of Indonesia's health budget during 2010-2022 period based on state budget realization and projection data. The analysis was conducted on three main components of the health budget, namely central government expenditure through Ministries/Institutions, transfers to regions and village funds, and health financing. Based on the analysis results, the health budget experienced significant increase from IDR 29.89 trillion in 2010 to IDR 255.39 trillion in 2022, with the ratio to total state expenditure increasing from 2.9% to 9.4%. The Ministry of Health became the largest allocation recipient in the central government expenditure category, while the regional transfer component showed consistent growth through Special Allocation Fund (DAK) for Health and Operational Health Assistance (BOK). Drastic increases in health budget occurred during 2020-2022 period, caused by COVID-19 pandemic response through various special programs such as health budget reserves, earmarked General Allocation Fund (DAU), and village funds for COVID-19 response. The results of this research provide a comprehensive overview of the Indonesian government's commitment in the health sector and can serve as a basis for evaluating the effectiveness of health budget allocation in the future.
Detection of Mental Health Tendencies Using Naïve Bayes Based on Social Media Activity Jeremi Sibarani; Ratih Manalu; Dongan Parulian Hutasoit; Wilman Arif Telaumbanua; Victor Asido Elyakim P
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 4 No. 2 (2025): Juni 2025
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v4i2.5959

Abstract

The development of social media has had a significant impact on individual mental health. This study aims to detect mental health trends based on user activity on social media using the Naïve Bayes algorithm. The data used is sourced from the Kaggle platform and collected through web scraping techniques with keywords related to mental health and social media activity. The analysis process includes data preprocessing, classification using Naïve Bayes, and evaluation of model performance by dividing training and test data at a ratio of 60:40, 70:30, and 80:20. The results showed that the Naïve Bayes method was able to classify mental health tendencies with the highest accuracy of 75.17% at a ratio of 60:40. Precision and recall were higher for the “Troubled” category compared to the “Good” category, showing the effectiveness of the model in detecting indications of mental disorders. However, there is still a prediction imbalance that affects the overall accuracy. These findings suggest that the Naïve Bayes algorithm can be a tool in social media-based mental health early detection, which can be used by health practitioners and researchers to design more appropriate intervention strategies.
Application of the Gaussian Mixture Models (GMM) Algorithm to Identify Error Patterns in Compilation Ayu Utari Nasution; Eko Prima Ambarita; Nurhidayanti Nurhidayanti; Heba Elsisy Fadlia; Victor Asido Elyakim P
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 4 No. 2 (2025): Juni 2025
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v4i2.5960

Abstract

In software development, the compilation process is a critical step that transforms source code into executable programs. These compilation errors can vary from simple syntax errors to more complex semantic errors, which often require a lot of time and effort to identify and fix. As a result, identifying recurring error patterns in the compilation process is important to improve software development efficiency. This study aims to explore the application of GMM in identifying error patterns in the compilation process. The results of this study indicate that the value of 0.58 on the Silhouette Score indicates that the clustering performed by GMM is quite good at identifying error patterns in compilation. The clusters are divided into 3, namely, Cluster 0 may indicate types of errors that occur more quickly (possibly related to syntax errors), with fewer lines of code and lower error frequency. Cluster 1 may represent more complex and less frequent errors (e.g., linker or runtime errors), with more lines of code. Cluster 2 may contain errors with different patterns, such as higher compilation duration or more frequent error frequency.
Annual Rainfall Prediction in Indonesia Using A Hybrid Artificial Neural Network and Fuzzy Algorithm Model Siti Asiah; Wanda Riana; Dika Chryston Purba; M Ilham Azharsum; Victor Asido Elyakim P
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 4 No. 2 (2025): Juni 2025
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v4i2.5964

Abstract

Rainfall is an essential meteorological parameter that affects various sectors of life. Accurately predicting rainfall has become crucial, and artificial intelligence-based models are increasingly popular in this field. Artificial Neural Networks (ANNs) have been widely used due to their ability to identify non-linear patterns in complex data. However, ANN-based predictions have limitations in optimally handling uncertainty or data variability. To address this issue, this study proposes a hybrid model that combines ANNs with fuzzy algorithms. Fuzzy algorithms are capable of managing uncertainty and providing flexible decision-making. This research proposes a hybrid model that integrates Artificial Neural Networks (ANNs) and fuzzy algorithms to predict annual rainfall based on meteorological data from 2019 to 2024. ANNs are used to detect non-linear patterns in temperature, humidity, and atmospheric pressure data, while fuzzy algorithms handle the uncertainty in input data. The model was tested using data from local meteorological stations and evaluated using MAE, RMSE, and the coefficient of determination (R²) metrics. The evaluation results show that the hybrid model achieved the best performance, with an MAE of 3.17 mm, RMSE of 3.4 mm, and R² of 0.98. These findings indicate that the combination of ANN and fuzzy logic significantly improves the accuracy of rainfall prediction compared to individual methods. This model has the potential to be applied in early warning systems and more precise climate management.
Co-Authors Adinda Febiola Ahmad Ryandika Aklima Laduna Ramadya Alisa Putri Amanda Nasution Andrea Erich Aprillya Zahra Iswandy Lubis Auralia Izmi Ayu Utari Nasution Bagas Adi Nata Baginda Faustine Sinaga Daniel Desmanto Sihombing Deswita Maharani Harahap Dian Yusri Andira Dika Chryston Purba Dimas Prayogi Dongan Parulian Hutasoit Ega Wahyu Andani Eko Prima Ambarita Fahriya Ardiningrum Ferdy Prayogi Fernando Siahaan Fransco Immanuel G Siahaan Haya Atiqah Tampubolon Heba Elsisy Fadlia Indah Frian Sari Jeremi Sibarani M Arif Y Sinaga M Ilham Azharsum M.Alfathan Haris Michael Orlando A. Purba Mirza Afif Pradivta Muhammad Rizky Ramadhan Nurhidayanti Nurhidayanti Olivia Diwani Saragih P.A.M. Zidane R.W.P.P. Zer Permaisuri Siregar Rahma Dhea Safitri Ratih Manalu Regita Audyna Siregar Rizky Nurhasanah Sella Ujani Siti Asiah Tiara Dwi Lestari Purba Vibiola Br Damanik Wanda Riana Wilman Arif Telaumbanua Yayang Arum Kemangi Yuegilion Pranayama Purba