cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Muhammad Syahrizal
Contact Email
syahrizal83.budidarma@gmail.com
Phone
+6282370070808
Journal Mail Official
pdsi.bids@gmail.com
Editorial Address
Jalan sisingamangaraja No 338 Medan, Indonesia
Location
Kota medan,
Sumatera utara
INDONESIA
Bulletin of Informatics and Data Science
Published by Asosiasi Peneliti Data Science Indonesia
ISSN : -     EISSN : 25808389     DOI : -
Core Subject : Science, Engineering,
Computer Science & IT Electrical & Electronics Engineering Engineering
The Bulletin of Informatics and Data Science journal discusses studies in the fields of Informatics, DSS, AI, and ES, as a forum for expressing research results both conceptually and technically related to Data Science
Arjuna Subject : Ilmu Komputer (semua kategori) - Kecerdasan Buatan
Articles 41 Documents
 1   2   3   4   5 
Implementation of XGBoost Ensemble and Support Vector Machine For Gender Classification of Skull Bones Ramadhani, Astrid; Afrianty, Iis; Budianita, Elvia; Gusti, Siska Kurnia
Bulletin of Informatics and Data Science Vol 4, No 1 (2025): May 2025
Publisher : PDSI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61944/bids.v4i1.115

Abstract

Sex identification based on skull bones is an important step in forensic anthropology, especially in cases where unidentified human skeletons are found. Conventional methods such as DNA analysis are often used, but have limitations, especially when the bones are damaged, charred or decayed, making the analysis process difficult. This research applies XGBoost ensemble and Support Vector Machine for sex classification on skull bones. The purpose of this research is to handle complex data with many features and unbalanced data using the XGBoost ensemble method and Support Vector Machine (SVM). The data used consisted of 2,524 samples with 82 measurement features. Model performance was evaluated using accuracy, precision, recall, and F1 score metrics. The results showed that the combination of XGBoost and SVM methods, especially with the RBF kernel, was able to achieve accuracy of up to 91.52%. This finding proves that machine learning-based approaches can be an effective and reliable solution in supporting the forensic identification process
Selection of the Best Customer using a Combination of Rank Order Centroid and Grey Relational Analysis Arshad, Muhammad Waqas; Rahmanto, Yuri; Setiawansyah, Setiawansyah
Bulletin of Informatics and Data Science Vol 3, No 1 (2024): May 2024
Publisher : PDSI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61944/bids.v3i1.84

Abstract

A customer is an individual or entity that purchases goods or services from a company or organization. They play an important role in business success, as customer satisfaction and loyalty can determine a company's reputation and sustainability in the marketplace. One of the main challenges is collecting and analyzing accurate and comprehensive data regarding purchase behavior, transaction frequency. Other challenges include keeping customer data confidential and ensuring that the selection process is fair and transparent. The ROC method is used in the initial stage to determine the importance weight of each criterion based on the subjective ranking of the decision makers, which is then converted into numerical weights systematically and consistently, the GRA method is applied to calculate the relational proximity between each customer's alternative to the ideal solution based on their performance values on each criterion.The purpose of this study is to develop and implement a comprehensive framework for the selection of the best customers by combining ROC weighting and GRA methods, and provide practical recommendations for companies in managing and utilizing the best customer relationships, in order to improve customer loyalty and long-term profitability. By combining these approaches, businesses can effectively prioritize customers based on their significance and potential to build long-term relationships and maximize profitability, thus enabling more targeted marketing strategies and better resource allocation. The best customer ranking results were obtained by Customer I with a final GRG value of 0.1792 for the 1st rank, Customer D with a final GRG value of 0.1683 for the 2nd rank, and Customer K with a final GRG value of 0.1505 for the 3rd rank
Hybrid Gradient Boosting and SMOTE-ENN for Toddler Nutritional Status Classification on Imbalanced Data Sinlae, Alfry Aristo Jansen; Erkamim, Moh.; Fitriyadi, Farid; Suhery, Lilik; Destriana, Rachmat
Bulletin of Informatics and Data Science Vol 3, No 2 (2024): November 2024
Publisher : PDSI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61944/bids.v3i2.93

Abstract

Stunting in toddlers remains a serious global health issue with long-term impacts on children's physical and cognitive development. One of the main challenges in classifying nutritional status is class imbalance, where the number of normal cases significantly exceeds that of minority classes such as stunted and severely stunted. This study aims to develop a hybrid approach by integrating the Gradient Boosting algorithm with the SMOTE-ENN (Synthetic Minority Oversampling Technique–Edited Nearest Neighbors) method to improve classification performance on imbalanced data. The dataset used was obtained from the Kaggle platform, consisting of 121,000 entries with four nutritional status categories. Data preprocessing included label encoding, numerical feature standardization, and stratified data splitting with an 80:20 ratio. The model was evaluated using accuracy, precision, recall, F1-score, and ROC-AUC metrics. The proposed hybrid model successfully increased the recall for the stunted class from 61.80% to 98.41%, and the F1-score from 71.93% to 83.58%. Overall accuracy improved from 92.39% to 93.35%, while the ROC-AUC score increased from 99.08% to 99.63%. These findings demonstrate that integrating Gradient Boosting with SMOTE-ENN is effective in enhancing sensitivity to minority classes and improving overall multi-class classification performance.
Decision Support System for Poverty Social Assistance using SMART and AHP Septhian, Haniel Dwi; Supriyanto, Aji
Bulletin of Informatics and Data Science Vol 4, No 1 (2025): May 2025
Publisher : PDSI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61944/bids.v4i1.106

Abstract

Poverty is a multidimensional problem that requires prompt and appropriate handling to maintain a dignified human life. In Manyaran Sub-district, Semarang City, the distribution of social assistance often faces obstacles due to limited human resources and a manual selection process for recipients. Therefore, a Decision Support System (DSS) is needed to assist the selection process in a more objective and efficient manner. This study aims to develop a DSS for determining social assistance recipients in Manyaran Sub-district by combining the Simple Multi-Attribute Rating Technique (SMART) and Analytical Hierarchy Process (AHP) methods. AHP is utilized to determine the weight of each criterion, while SMART is used to calculate the final score of each recipient candidate. The combination of SMART and AHP allows for both expert-based prioritization and quantitative evaluation, enhancing transparency and consistency in the selection process. The research was conducted through stages of problem analysis, data collection, literature review, system design, and report writing. The results show that among the ten analyzed candidates, the individual coded P06 achieved the highest final score of 0.574. The top five candidates with the highest scores were declared eligible to receive social assistance, while the others were declared ineligible. The application of the SMART and AHP methods in this DSS effectively improves the accuracy, objectivity, and efficiency of the selection process for social assistance recipients in Manyaran Sub-district
Effectiveness of Weighting in Assessing Ranking Criteria on the SWOT-MAGIQ Matrix Ambarsari, Erlin Windia; Subagio, Relo; Mesran, Mesran
Bulletin of Informatics and Data Science Vol 3, No 1 (2024): May 2024
Publisher : PDSI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61944/bids.v3i1.85

Abstract

The Analytical Hierarchy Process (AHP) has been a prominent tool in decision-making, but the Multi-Attribute Global Interference of Quality (MAGIQ) offers an alternative with its unique weighting mechanism. This research delves into the effectiveness of weighting in assessing ranking criteria within the SWOT-MAGIQ matrix. The study contrasts the traditional Rank Order Centroid (ROC) approach with the Improved Rank Order Centroid (IROC), focusing on their application in the SWOT analysis. While ROC provides simplicity, IROC aims for enhanced accuracy by considering variability in rankings. The results indicate nuanced differences, with ROC assigning higher weights to criteria such as "Friendly Staff" (0.3183 vs. IROC’s 0.3125), while IROC prioritizes aspects like "Strong Customer Relationships" more significantly (0.1103 vs. ROC’s 0.1053). The choice between ROC and IROC hinges on the specific needs of the decision-making context, with IROC potentially offering a more detailed perspective in complex scenarios. This research underscores the importance of selecting the appropriate weighting mechanism to ensure informed and strategic decisions within the SWOT-MAGIQ framework
Enhancing Support Vector Machine Performance for Heart Attack Prediction using RobustScaler-Based Outlier Handling Lasiyono, M Munawir; Nurhayati, Nurhayati; Soares, Teotino Gomes; Mulyadi, Mulyadi
Bulletin of Informatics and Data Science Vol 4, No 1 (2025): May 2025
Publisher : PDSI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61944/bids.v4i1.94

Abstract

Cardiovascular disease remains the leading cause of death worldwide, with most cases attributed to heart attacks and strokes. Early detection is crucial, yet conventional diagnostic methods are often constrained by time, cost, and uneven distribution of clinical expertise. Consequently, machine learning-based approaches offer a promising alternative for efficiently supporting heart attack prediction. This study employs the Support Vector Machine (SVM) algorithm, focusing on enhancing its performance through RobustScaler as a preprocessing technique to address outliers common in medical datasets. The objective of this study is to evaluate the impact of RobustScaler on SVM performance in heart attack classification. The model was developed using a dataset of 303 patient records, consisting of eight numerical features and one binary target label. Experiments were conducted under two preprocessing scenarios: without scaling (baseline) and with RobustScaler. Model performance was assessed using accuracy, precision, recall, F1-score, and ROC-AUC. The results show that applying RobustScaler significantly improves model performance, with accuracy increasing from 64.77% to 85.23%, representing a 20.46% improvement, and ROC-AUC rising from 73.65% to 93.36%, indicating a 26.78% increase in discriminatory ability. Additionally, recall for the negative class improved dramatically from 26.47% to 99.02%, reflecting better sensitivity in identifying non-heart attack cases. These findings demonstrate that proper preprocessing, particularly using RobustScaler, plays a vital role in optimizing SVM performance, especially when handling clinical data with extreme values
Decision Support System for Selecting the Best Head of Study Program Applying the Multi-Objective Optimization Method on the Basis of Simple Ratio Analysis (MOOSRA) Lubis, Juanda Hakim; Sanwani, Sanwani; Lubis, Akhyar; Mesran, Mesran; Julaysa, Julaysa; Hutapea, Novita Sari
Bulletin of Informatics and Data Science Vol 4, No 1 (2025): May 2025
Publisher : PDSI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61944/bids.v4i1.105

Abstract

The Head of Study Program plays an important role as the highest leader in the structure of a study program. The Head of Study Program is responsible for the smooth running of academic activities in the study program he leads. As a key element in higher education, the Head of Study Program must lead the managerial function by planning, implementing, and controlling the academic process and managing other supporting activities. The head of the study program who shows high performance, dedication, and integrity deserves an award as the best head of the study program. This assessment aims to ensure that the Head of Study Program is able to carry out his duties properly in accordance with the rules and demands, and advance the study program in accordance with its vision and mission. Therefore, a decision support system is needed as a solution to overcome this problem, by utilizing the MOOSRA method. MOOSRA begins by formulating a decision matrix consisting of alternatives, criteria or attributes, individual weights or significance coefficients of each criterion, and performance measurements of related alternatives. Normalization is then carried out to change the attribute values into the range 0–1. The assessment results show that the Head of Study Program with the highest ranking is alternative A7, with a value of 0.896358
Implementation of Decision Support System in Choosing the Best E-Wallet using ROC and MOORA Weighting Methods Mesran, Mesran; Qomaini, Ahmad; Sirait, Ika Paulina; Rosnizam, Rosnizam
Bulletin of Informatics and Data Science Vol 3, No 1 (2024): May 2024
Publisher : PDSI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61944/bids.v3i1.87

Abstract

This research aims to analyze digital wallet usage preferences among university students using a decision support system. With the rise of various digital wallet services such as DANA, LinkAja, and GoPay, students often have difficulty in choosing the platform that best suits their needs. To solve this problem, the research uses a combination of the Rank Order Centroid (ROC) and Multi-Objective Optimization by Ratio Analysis (MOORA) methods. The ROC method was used to determine the criteria weights, while MOORA was applied to rank the digital wallet alternatives. The criteria used in this study include merchant range, cashback program, transaction fees, and ease of use. The research sample consisted of 150 students from the Faculty of Economics, Universitas Nusantara. The results showed that GoPay ranked first as the most preferred digital wallet, followed by DANA in second place, and LinkAja in third. The findings are expected to help students in choosing the digital wallet that best suits their preferences, as well as provide input for digital wallet service providers to improve their service quality
Diabetes Classification using Gain Ratio Feature Selection in Support Vector Machine Method Al Rasyid, Nabila; Afrianty, Iis; Budianita, Elvia; Kurnia Gusti, Siska
Bulletin of Informatics and Data Science Vol 4, No 1 (2025): May 2025
Publisher : PDSI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61944/bids.v4i1.114

Abstract

Diabetes is a major cause of many chronic diseases such as visual impairment, stroke and kidney failure. Early detection especially in groups that have a high risk of developing diabetes needs to be done to prevent problems that have a wide impact. Indonesia is ranked seventh in the world with a prevalence of 10.7% of the total number of people with diabetes. This research aims to determine the attributes in the diabetes dataset that most affect the classification and apply the Support Vector Machine method for diabetes classification. For the determination process, Gain Ratio feature selection technique is applied. The dataset used consists of 768 data with 8 attributes. In this classification process, 3 SVM kernels (Linear, Polynomial, and RBF) are used with three possible data divisions using the ratio (70:30; 80:20; 90:10). Before applying feature selection, there were 8 attributes used and achieved the highest accuracy of 94.81% at a ratio of 80:20 using the RBF kernel with a combination of two parameters namely C = 100, Gamma = 3 and C = 100, Gamma = Scale.  Feature selection parameters in the form of thresholds used include 0.02; 0.03; and 0.05. After applying feature selection, the attribute that produces the highest accuracy uses 6 attributes. The highest accuracy after applying feature selection reached 95.45% at a threshold of 0.02 with a ratio of 80:20 using the RBF kernel with parameters C = 100 and Gamma = Scale. The results showed that there was an increase in accuracy after applying feature selection
Optimizing Autoencoder-Based Feature Selection for Attack Detection in IoT Networks via Machine Learning Approaches Winanto, Eko Arip; Kurniabudi, Kurniabudi; Sharipuddin, Sharipuddin
Bulletin of Informatics and Data Science Vol 3, No 2 (2024): November 2024
Publisher : PDSI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61944/bids.v3i2.104

Abstract

The Internet of Things (IoT) presents significant security challenges as the number of connected devices continues to grow. One critical approach in developing efficient attack detection systems is the selection of relevant features to reduce model complexity without compromising accuracy. This study evaluates the effectiveness of Autoencoders as a feature reduction method for IoT network intrusion detection systems. Three machine learning algorithms are employed for comparative analysis: K-Nearest Neighbors (KNN), Naive Bayes (NB), and Support Vector Machine (SVM). The dataset is evaluated both before and after feature reduction using an Autoencoder, with performance assessed based on accuracy, precision, recall, F1-score, training time, and the number of features. Experimental results demonstrate that the Autoencoder can reduce the number of features by up to 30% without significantly degrading performance. In fact, the NB and SVM models exhibit improvements in both accuracy and training efficiency. The KNN model shows a minimal performance decline, which remains within acceptable limits. Overall, the Autoencoder proves to be an effective method for feature reduction, maintaining or even enhancing detection efficiency and performance. These findings support the use of Autoencoders as an efficient feature selection technique in IoT-based attack detection systems.

Page 4 of 5 | Total Record : 41

 1   2   3   4   5 

Filter by Year

2022 2025


Reset
Filter By Issues
All Issue Vol 4, No 1 (2025): May 2025 Vol 3, No 2 (2024): November 2024 Vol 3, No 1 (2024): May 2024 Vol 2, No 2 (2023): November 2023 Vol 2, No 1 (2023): May 2023 Vol 1, No 2 (2022): November 2022 Vol 1, No 1 (2022): May 2022