Article Per Year (5 Year)
p-Index From 2021 - 2026
0.92
P-Index
This Author published in this journals
All Journal JOURNAL OF INFORMATION SYSTEM RESEARCH (JOSH) Bulletin of Computer Science Research
Sufajar Butsianto
Universitas Pelita Bangsa, Bekasi
Computer Science & IT Decision Sciences, Operations Research & Management

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

Found 5 Documents
Search

 1 
Implementasi Algoritma Apriori dalam Menemukan Pola Asosiasi pada Data Penjualan Produk Retail Sufajar Butsianto; Candra Naya; Anggi Muhammad Rifa'i
Bulletin of Computer Science Research Vol. 5 No. 5 (2025): August 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v5i5.731

Abstract

This study aims to implement the Apriori algorithm in finding association patterns in retail product sales data, using the Association Rule Mining approach. Evaluating the ruler or association rules formed based on the support, confidence, and lift parameters, in finding association patterns in retail product sales data with a focus on the relationship between product categories. The data used consists of 500 sales data as sample data and 5,972 transactions as test data. The data mining process was carried out on the main product categories such as Milk, Coffee, Tea, Drinks, Detergent, and Biscuit/Snacks, to find association rules that appear simultaneously with the Bulk Products category in one transaction time. The minimum support parameter was set at 0.02 and the minimum confidence was set at 0.5. By using these parameters, several significant association rules were obtained. One of the strongest rules shows that if products in the Milk, Coffee, Tea, Drinks, Detergent, and Biscuit/Snacks categories are purchased together, then there is a 64.3% probability (confidence) that products in the Bulk Products category are also purchased at the same time. The support value of this rule reached 3.8%, and the lift value was 1.49, indicating a positive association and not a coincidence. Evaluation of the test data showed that this pattern was consistently found across 5,972 transactions, with a repeatability rate of 61.7%. The results of this study demonstrate that the Apriori algorithm is effective in identifying consumer purchasing patterns that can be utilized for product placement strategies, bundling offers, and inventory planning in retail management.
Perbandingan Kinerja Naïve Bayes dengan dan Tanpa SMOTE untuk Klasifikasi Gangguan Kecemasan Mahasiswa pada Data Tidak Seimbang Nurhadi Surojudin; Sufajar Butsianto; Andri Firmansyah
Bulletin of Computer Science Research Vol. 6 No. 2 (2026): February 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v6i2.1021

Abstract

Anxiety disorders are one of the most common mental health problems experienced by university students and may affect learning concentration and academic performance. The analysis of psychological survey data using machine learning techniques can support early detection of student anxiety conditions. However, one of the main challenges in mental health data analysis is the presence of class imbalance within the dataset. This study aims to analyze the effect of applying the Synthetic Minority Oversampling Technique (SMOTE) on the performance of the Naïve Bayes algorithm for multi-class classification of student anxiety levels, which are categorized into three classes: No Stress, Eustress, and Distress. The dataset used in this research was obtained from student questionnaire data and underwent several preprocessing steps including data cleaning, feature transformation, and dataset splitting using the hold-out method with a ratio of 80% training data and 20% testing data. Model performance was evaluated using a confusion matrix with evaluation metrics including accuracy, precision, recall, and F1-score. The results show that the Naïve Bayes model without SMOTE achieved an accuracy of 0.84, precision 0.78, recall 0.41, and F1-score 0.54. After applying SMOTE, the model achieved an accuracy of 0.82, precision 0.74, recall 0.69, and F1-score 0.71. These results indicate that SMOTE improves the model's ability to detect minority classes in multi-class classification problems, although a slight decrease in overall accuracy is observed.
Analisis Sentimen Pengguna X Terhadap Isu Adzan Menjadi Running Text Menggunakan Algoritma K-Nearest Neighbors (KNN) Arya Rifaldi Baharudin; Sufajar Butsianto; Asep Supriyanto
Journal of Information System Research (JOSH) Vol 6 No 4 (2025): July 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v6i4.8055

Abstract

Sentiment analysis has become an important method for understanding public opinion on social and religious issues. This study aims to analyze user sentiment regarding the issue of the adzan presented in a running text format using the K-Nearest Neighbors (K-NN) algorithm. The adzan as running text on national television occurred during Pope Francis's mass at Gelora Bung Karno Stadium (GBK) on September 5, 2024. The Ministry of Religious Affairs (Kemenag) advised that the Maghrib adzan, usually broadcast on national television, be replaced with running text. This recommendation was made to facilitate the live broadcast of the mass attended by Christian congregants and to honor the worship without disruption. Some parties, such as the Indonesian Ulema Council (MUI) and the General Chairman of PP Persis, stated that replacing the adzan with running text does not violate Islamic law, while Minister of Communication and Information Budi Arie Setiadi mentioned that the change is merely a suggestion. The research findings indicate that the K-NN algorithm involves several stages, including data collection and labeling, text processing, feature extraction using TF-IDF, and splitting the data into 80% for Training and 20% for Testing. Based on the test results, the K-NN model detected 10 positive sentiments and 168 negative sentiments, indicating a tendency for Twitter users to express more negative sentiments. Analysis using a Confusion matrix shows that this model achieved an accuracy rate of 88%, indicating good performance in sentiment classification.
Implementasi K-Means Clustering Berbasis RapidMiner untuk Optimalisasi Segmentasi Penjualan Produk dalam Meningkatkan Efektivitas Strategi Pemasaran Sufajar Butsianto; Arif Siswandi
Journal of Information System Research (JOSH) Vol 7 No 1 (2025): October 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v7i1.8439

Abstract

The Indonesian electronic retail industry is experiencing rapid growth along with digital transformation. However, available sales data is often only stored as transaction records without further analysis, so it has not been optimally utilized for marketing decision making or customer segmentation. This study aims to implement the RapidMiner-based K-Means Clustering algorithm to analyze segmentation patterns of electronic products at XYZ Store. The dataset used includes the variables Transaction_ID, Product_ID, Product_Name, Category, Quantity, Unit_Price, Revenue, and Recency. The research stages include data collection, preprocessing (filtering, aggregation, and Z-Score normalization), K-Means application, and interpretation of clustering results. Determination of the number of clusters in this study uses the Elbow Method, which shows an optimal point at K = 3, so that number of clusters is chosen for the data grouping process. Based on the results of the application of the K-Means algorithm with the three clusters, the following are obtained: (1) Cluster 0 (High Sales & High Revenue) consisting of Smartphones, Laptops, and Tablets as superior products with a contribution of almost 60% of total revenue; (2) Cluster 1 (Medium Sales & Moderate Revenue) includes Televisions, Refrigerators, and Smartwatches with a stable contribution of around 27%; and (3) Cluster 2 (Low Sales & Low Revenue) contains Washing Machines, Speakers, Headphones, and Printers with a low contribution of only 14%. These findings provide a strategic basis for management in making business decisions, such as procurement priorities, seasonal promotions, product bundling, and clearance strategies. This study proves that the application of data mining with K-Means Clustering is effective in increasing operational efficiency and supporting the competitiveness of the electronics retail business in Indonesia.
Prediksi Risiko Penyakit Jantung Menggunakan Support Vector Machine dengan Seleksi Fitur dan Optimasi Hyperparamete Nurhadi Surojudin; Sufajar Butsianto; Rizal Ainun Yaqin
Journal of Information System Research (JOSH) Vol 7 No 4 (2026): July 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v7i4.10518

Abstract

Heart disease remains one of the leading causes of mortality worldwide, highlighting the need for accurate prediction models to support early diagnosis and clinical decision-making. This study proposes a heart disease risk prediction model based on Support Vector Machine (SVM) integrated with Recursive Feature Elimination (RFE) for feature selection and GridSearchCV for hyperparameter optimization. The study utilized the Cleveland Heart Disease Dataset, consisting of 303 patient records, 13 predictive attributes, and one target variable. The research workflow included dataset collection, data preprocessing, feature selection, hyperparameter optimization, model development, and performance evaluation using Accuracy, Precision, Recall, F1-score, and ROC-AUC. Experimental results demonstrate that the proposed model achieved an Accuracy of 86.89%, Precision of 85.71%, Recall of 85.71%, F1-score of 85.71%, and ROC-AUC of 93.18%. Feature selection successfully reduced irrelevant attributes, improving model efficiency, while hyperparameter optimization produced a more effective parameter configuration than the default settings. These findings indicate that integrating RFE with GridSearchCV enhances the predictive performance of SVM and provides a promising approach for supporting heart disease diagnosis using machine learning techniques.
Co-Authors Andri Firmansyah Anggi Muhammad Rifa'i Arif Siswandi Arya Rifaldi Baharudin Asep Supriyanto Candra Naya Nurhadi Surojudin Rizal Ainun Yaqin