Jurnal Sistemasi (OJS FTIK - UNISI, Fakultas Teknik dan Ilmu Komputer Universitas Islam Indragiri)
Not a member yet
946 research outputs found
Sort by
Web-based Educational Payment Information System using Role-based Access Control Security
Get PDFThe management of Sumbangan Pembinaan Pendidikan (SPP) payments in Islamic boarding schools still faces issues related to delayed recording, data inaccuracies, and weak access control over financial information. This study aims to design and implement a web-based SPP payment information system that applies Role-Based Access Control (RBAC) to improve administrative order and data security. The system was developed using the Waterfall method, which consists of requirement analysis, system design, implementation, testing, and maintenance stages. The application was built using the Laravel framework with RBAC implemented at the middleware level to manage user access based on defined roles, namely Super Admin, Treasurer, Student Guardian, and Principal. System testing was conducted using the Black-Box Testing method to validate core functionalities, including user authentication, billing management, payment verification, report generation, and role-based access restrictions. The test results indicate that all system functions operate as expected and that the RBAC mechanism effectively prevents unauthorized access to sensitive features and data. Overall, the implemented system supports more structured payment administration, improves data accuracy, and enhances security and accountability in managing financial transactions within the pesantren environment
Design of a Palm Oil Harvest Recording Information System using Two-Factor Authentication Security
Get PDFThe palm oil harvest recording process in many farmer groups is still carried out manually, which may lead to recording errors, delays in reporting, and difficulties in verifying harvest data. This study aims to design and develop a web-based palm oil harvest recording information system that supports harvest data entry by farmers, harvest verification by agents, transportation monitoring by drivers, and the presentation of harvest information through a dashboard for the owner. The system is designed by integrating a Two-Factor Authentication (2FA) security mechanism using One-Time Password (OTP) verification as an additional authentication layer to enhance user access security. The software development method applied in this study is the System Development Life Cycle (SDLC) Waterfall model, which includes requirement analysis, system design, implementation, testing, and maintenance stages. Black-box testing results indicate that all main system features function properly according to user requirements. In addition, user acceptance testing using usability testing with the System Usability Scale (SUS) instrument obtained an average score of 72.5, which falls into the Good category. These results suggest that implementing OTP-based 2FA improves authentication security without reducing system usability. The developed system is able to enhance the efficiency, accuracy, and transparency of the palm oil harvest recording process and has the potential to be implemented in farmer groups or similar business units
Analysis of Cryptocurrency Candlestick Patterns using Gramian Angular Field and Hybrid Deep Learning
Get PDFCryptocurrency markets such as Bitcoin, Ethereum, and Solana exhibit high volatility, making price forecasting difficult when relying solely on conventional technical analysis. This study aims to analyze cryptocurrency candlestick patterns by utilizing Gramian Angular Field (GAF) representations and to evaluate the performance of a hybrid deep learning model combining CNN–LSTM–Transformer to support investment decision-making. The proposed method involves processing daily historical Open, High, Low, and Close (OHLC) data from three major cryptocurrency assets: Bitcoin (BTC-USD), Ethereum (ETH-USD), and Solana (SOL-USD), covering the period from January 1, 2020, to September 30, 2024, obtained from Yahoo Finance. The time-series data were transformed into 64×64 pixel GAF images and used to train a baseline CNN model as well as a hybrid CNN–LSTM–Transformer model. Model evaluation was conducted across multiple forecasting horizons, including 1 day, 7 days, 30 days, 180 days, and 1 year, and was further complemented by real-time testing using the CoinGecko API in March 2025. The results indicate that the hybrid model achieved the best performance at different horizons for each asset: BTC-USD at the 30-day horizon with an R² of 0.971 and an SMAPE of 0.77%, ETH-USD at the 1-year horizon with an R² of 0.948 and an SMAPE of 0.81%, and SOL-USD at the 1-year horizon with an R² of 0.910 and an SMAPE of 4.72%. Real-time testing demonstrated that the model consistently captured the overall price movement trends despite high market volatility. It can be concluded that the integration of GAF representations and the hybrid CNN–LSTM–Transformer model has strong potential to enhance cryptocurrency candlestick analysis and can be utilized as a component of a Decision Support System for digital asset investment
Literature Review: Comparison of Machine Learning Algorithms for Sentiment Analysis of Free Nutritious Meals
Get PDFThe Free Nutritious Meal (FNM) program has triggered massive public responses on social media, driving numerous machine learning–based sentiment analysis studies. However, there has been no comprehensive review comparing the effectiveness of these methods. This study adopts a Systematic Literature Review (SLR) approach on 18 studies (2024–2026) to evaluate the performance of computational algorithms and map trends in public sentiment. The main contribution of this research is to provide an empirical guide for selecting Indonesian-language text classification models, while also offering insights into shifts in public perception. Key findings indicate that Support Vector Machine (SVM) is the most frequently used method, whereas the highest accuracy (97%) was achieved by a combination of Logistic Regression, SVM, and Random Forest on large datasets. Temporally, sentiment trends shifted from budget skepticism (2024) to positive acceptance during program implementation (2025–2026). The study’s implications support policymakers in evaluating program effectiveness in real time. The scope and limitations of this research focus on literature within a specific timeframe, with performance evaluation emphasizing quantitative accuracy metrics
Digital Image Confidentiality using New Encryption Method
Get PDFThe significant development in mobile phone cameras, in terms of the clarity and high-resolution images captured, the wide utilizing of mobile phones and other communication devices among all segments of society, and the increasing use of social networking sites and the exchange of millions of digital photos daily, all have been pointed with a great importance of digital images and the need to provide adequate security and protection for these images. Digital images contain a large amount of information and recently they have an effective and easy tools of communication without the need for a determined text.Given that digital images contain important, personal, and sensitive information, there is a great need to protect these images and prevent unauthorized persons from applying any changes to the image's contents. There is a great deal of work in this field, most of which uses encryption methods to achieve this protection. As is well known, there are two types of encryption systems (symmetric and asymmetric). Symmetric encryption systems are fast but require a secret key distribution process, while asymmetric encryption systems are relatively slow, involving complex processes. Therefore, they are not suitable for use in social networking applications that require rapid performance and interaction. In this research, a proposed method for digital image encryption is proposed, which includes the use of logical XOR operation to encrypt the digital image based on two proposed levels with scrambling operation to provide a high degree of diffusion and confusion for the resulting encrypted image. The proposed method was evaluated through a set of efficiency measurement metrics (NPCR, UACI, MSE, PSNR, SSIM, Entropy, and Correlation) and it gave results showed a difference between the original image and the image resulting from the first level of encryption. We also noted that the image resulting from the second level had a higher percentage of difference and randomness compared to the original image. Therefore, the proposed encryption method is suitable in terms of speed and confidentiality for use in encrypting digital images and thus maintaining their privacy
Predict Airline Customer Satisfaction using a Machine Learning Model
Get PDFCustomer satisfaction is a strategic factor for the sustainability of airline businesses amid increasingly intense competition in the aviation industry. This study aims to predict airline customer satisfaction using an Artificial Neural Network (ANN) approach by leveraging a publicly available Kaggle dataset containing 22 airline service features. Two ANN architectures were developed, differing primarily in the number of hidden layers, the number of neurons, and the application of Batch Normalization and LeakyReLU in the second model. The experimental results show that the first ANN model achieves an accuracy of 92.31%, while the second model attains significantly higher performance, with an accuracy of 95.75% on the test dataset. The second model also demonstrates a strong balance between precision and recall (0.94–0.97), with an average F1-score of 0.95–0.96 and a minimal number of misclassifications. These results confirm that employing a more complex ANN architecture can deliver highly accurate predictions of customer satisfaction. The implementation of ANN-based predictive models not only enhances passenger experience quality but also strengthens customer loyalty and helps airlines maintain long-term competitiveness
Prototype of IoT-based Gas and Temperature Monitoring System for Genset Room with ESP8266
Get PDFGenerator rooms require strict environmental monitoring to ensure operational safety, as uncontrolled temperature and gas concentration conditions may pose risks of fire and workplace accidents. This study aims to develop an Internet of Things (IoT)–based generator room monitoring system capable of automatically and real-time monitoring temperature, humidity, and gas levels, with notifications delivered via a Telegram bot. The research method adopts a systematic approach through prototype development. The system utilizes DHT22 and MQ-2 sensors, a NodeMCU ESP8266 microcontroller, a buzzer, and is integrated with the Telegram Bot API. The testing results demonstrate that the device is able to responsively monitor environmental parameters, transmit real-time data to Telegram, activate the relay with a response time of 4–5 seconds, and provide automatic notifications when gas concentrations exceed safe thresholds with a delay of 1–2 seconds. Evaluation of the monitoring functionality and sensor data visualization through the Telegram application indicates that all sensor information is successfully displayed via the Telegram bot interface. Although data transmission from the sensors to the Telegram application is highly dependent on network connection stability, the experimental results show that the system is capable of delivering data with optimal response time
Comparative Analysis of Naive Bayes and Fuzzy Logic Algorithm in Fire Classification System
Get PDFBuilding fires can cause losses in several areas, including property damage, environmental pollution, loss of life, injury, and psychological trauma. Building fires can occur due to several factors such as gas leaks, short circuits, overheating electronic devices, the presence of flammable materials, and human error. In fire mitigation efforts, devices are generally used as early warnings, but their implementation is often less than optimal due to system malfunctions. Therefore, this study aims to develop an early warning system that can detect potential fires before they spread. The methods used in this study are the naïve Bayes and fuzzy logic methods, which then compare each method to determine the most effective method. The results of this study indicate that the naïve Bayes and fuzzy logic methods have successfully classified potential fires well. From 30 experimental data, the naïve Bayes algorithm produced an accuracy of 96%, while the fuzzy logic algorithm produced an accuracy of 100%. The naïve Bayes algorithm shows reliable performance in classifying extreme data while the fuzzy algorithm can detect the ‘Danger’ status even though not all parameters are in a dangerous condition
Aspect-based Sentiment Analysis: A Bibliometric Review using Bibliometrix to Map Research Trends and Algorithm Methods
Get PDFThis study presents a bibliometric overview of research trends and algorithmic models in Aspect-Based Sentiment Analysis (ABSA). Data were collected from the Scopus database, resulting in a dataset of 2,344 journal articles published between 2021 and early 2026. The analysis was conducted using the Bibliometrix and Biblioshiny packages in R to perform number of publications per year, source’s production over time, country production over time, keyword co-occurrence, thematic mapping and evolution of research themes. The results show that ABSA research has experienced rapid growth with an annual publication increase of more than 30%. This study identifies BERT algorithmic models and Graph Convolutional Networks (GCN) as the most dominant supporting tools in the research literature. Thematic maps show that transformer-based techniques and attention mechanisms have emerged as key driving themes in this field. Furthermore, thematic evolution maps reveal a shift in focus from technical aspect extraction to online public opinion analysis, reinforced by the sharp surge in the use of Large Language Models (LLMs) in recent years. The findings provide a structured overview of the intellectual landscape of ABSA, clarifying dominant research clusters, methodological trajectories, and emerging themes. By highlighting the central role of transformer architectures, graph-based neural networks, and LLM integration, this study offers methodological guidance for future model development. Furthermore, the bibliometric insights reduce research fragmentation and identify underexplored directions, offering valuable insights for researchers to identify research gaps and develop more advanced ABSA models in future studies
Implementation of the Apriori Algorithm for Clothing Store Product Recommendations based on Sales Transaction History
Get PDFThis study is motivated by the limitations faced by small-scale clothing stores, which generally do not have customer ratings or reviews that can be used as a basis for product recommendations. This condition necessitates an alternative method capable of utilizing available sales transaction data. The objective of this study is to generate product recommendations by identifying consumer purchasing patterns through the application of the Apriori Algorithm. The methodology involves processing sales transaction data consisting of transaction codes, lists of purchased products, and transaction timestamps. Support, confidence, and lift ratio values are calculated to generate and validate association rules among products. The analyzed data are derived from the transaction history of a clothing store and are processed using a web-based system developed with PHP and MySQL. The experimental results indicate that several product combinations achieve confidence values of 50% and lift ratios greater than or equal to 1, suggesting that these patterns can be used as a basis for product recommendations. These findings demonstrate a strong association among items that are frequently purchased together. Based on the results, this study concludes that the Apriori Algorithm is effective in identifying meaningful purchasing patterns that can support product arrangement strategies and inventory management in small-scale clothing stores