SAGA: Journal of Technology and Information System

Implementation of K-Means Algorithm to Classify Instagram Reels and Carousel Content Performance Based

Shelomitha Trinitia Wowor — 2026-06-15

With the increasing popularity of digital marketing, Instagram became one of the top platforms where the audience can be reached. It is important to gain an insight into the performance of different types of contents to ensure that the marketing efforts bear fruit. This research will apply the K-Means algorithm to classify Instagram Reels and Carousel contents based on performance by taking into account such factors as likes, comments, shares, and saves. For the purposes of the study, the data were collected from a variety of accounts both personal and of a business nature. The number of clusters was defined by the Elbow Method, after which they were categorized depending on their performance such as high, medium, and low. The results indicate that the classification based on performance provided by the K-Means algorithm can provide insights into marketing practices on Instagram. Consequently, the present research will contribute to the development of digital marketing studies, particularly in the area of content analysis, within the field of data mining.

RadOnco-Priority: Machine Learning Decision Support for Radiotherapy Queue Prioritization Using Real-World Retrospective Radiotherapy Referral Data

Muhammad Sobri Maulana — 2026-07-21

Radiotherapy queues are often managed by referral date and manual clinician judgment, although limited linear accelerator capacity requires prioritization that is clinically transparent, operationally auditable, and fair. This study evaluates RadOnco-Priority, a machine learning-enabled decision support framework for radiotherapy queue prioritization, using a de-identified real-world retrospective dataset of 240 radiotherapy referral records rather than simulated or synthetic patient records. The system combines a literature-informed rule-based urgency score with supervised machine learning models to identify patients requiring accelerated booking. Accelerated booking need was defined a priori as an operational triage label reflecting clinician-documented priority, urgent symptoms, time-sensitive tumor-site and treatment-intent combinations, accumulated referral delay, and planning complexity. Logistic regression, random forest, and gradient boosting were trained to predict accelerated booking need, while a capacity-aware scheduling simulation evaluated waiting-time redistribution. To address potential circularity, additional ablation analyses were performed with the aggregate urgency score removed from the predictors. In the held-out test set, logistic regression achieved the highest discrimination in the full-feature model (AUC 0.91), with sensitivity-oriented classification favoring reduced false negatives. Performance remained acceptable after removing the aggregate urgency score, indicating that the model did not rely solely on the pre-specified scoring logic. The scheduling simulation reduced median waiting time in the high-priority group and decreased the proportion of high-priority patients waiting more than 28 days. These findings support RadOnco-Priority as an interpretable, human-governed information-system framework for radiotherapy queue management. Prospective multicenter validation, fairness monitoring, and local ethics approval remain required before routine implementation.

Raw Material Storage Information System Case Study: CV.Bplast Jaya

Aldo Ardiansyah — 2026-07-30

This study was motivated by the limitations of the raw material inventory management system in companies, which is still carried out manually. The lack of understanding and knowledge of information technology among employees is also a factor hindering the implementation of a more efficient system. This condition causes various obstacles, such as delays in recording, inaccurate data, and difficulties in real-time stock tracking. The purpose of this study is to assist companies in developing a more effective and efficient raw material storage system through the use of web-based information technology. This study uses a qualitative approach with data collection methods in the form of observation, interviews, and documentation to obtain information related to system requirements and problems faced by companies. Based on the results of the research, a web-based raw material storage information system has been successfully designed and developed that is capable of presenting real-time inventory data, recording the entry and exit of raw materials more accurately, and providing search and reporting features that facilitate decision-making by management. The implementation of this system has shown an increase in data recording efficiency, a reduction in input errors, and ease in tracking raw material stocks. Thus, the developed system can be the right solution to overcome the raw material management problems that the company has been facing.

Skrining Assist Information System v3.1 at the National Narcotics Agency of Sidoarjo Regency

Nella Prima Yeni — 2026-07-30

The Screening Assist Information System v3.1 is a web-based application designed to support the substance abuse screening process at the National Narcotics Agency (BNN) of Sidoarjo Regency. This system uses the ASSIST (Alcohol, Smoking and Substance Involvement Screening Test) method from the WHO as an early detection tool for substance use risks. The system was developed using a Research and Development (R&D) approach with the Waterfall model in the System Development Life Cycle (SDLC). This information system was created using the PHP programming language and Mysql database. With its time efficiency, scoring accuracy, and data security, this system is an effective digital solution in supporting efforts to prevent and combat drug abuse in the Sidoarjo Regency. The system evaluation was conducted using the Blackbox Testing method, which showed that all features functioned properly, and the User Acceptance Testing (UAT) method was used to measure user satisfaction with a result of 87.6% . This percentage reflects the expectation that students provided honest answers, so that the data obtained was close to the actual level of validity. Overall, the evaluation results indicate that the developed information system can serve as an effective, adaptive digital solution that supports professional and structured efforts to prevent and address drug abuse.

Model Analysis of the 2019 Election Participation Level Against Demographics in Pamekasan Regency Using the Naive Bayes Method

Maulana Habib Firmansyah — 2026-07-30

This study aims to analyze the level of election participation in 2019 in Pamekasan Regency based on demographic data using the Naive Bayes classification method. The data used consists of 189 instances and 208 predictor attributes obtained from the publication of the Central Statistics Agency (BPS). The analysis process involves the stages of preprocessing, feature selection, and model evaluation. The test results show a model accuracy of 66%, with the highest f1-score value in the high participation class. Further analysis also shows that most sub-districts and villages in Pamekasan have a high level of participation. In addition, a very strong correlation was found between demographic attributes that have the potential to be important predictors of voter involvement. These findings provide an initial overview to understand the factors that influence public participation in elections.