Journal of System and Computer Engineering

Implementation of Fisher-Yates Shuffle Algorithm in Mobile-Based Vocabulary Learning Game for Children with Disabilities

2026-05-05T14:09:50+00:00

Children with disabilities face significant challenges in vocabulary acquisition, necessitating the development of specialized educational technologies that accommodate their unique learning characteristics. This study aims to implement the Fisher-Yates shuffle algorithm in a mobile-based vocabulary learning game specifically designed for children with disabilities, ensuring unbiased randomization of educational content to promote authentic vocabulary comprehension. This research employed the Multimedia Development Life Cycle methodology, encompassing concept definition, design, material collection, assembly, testing, and distribution phases. The Fisher-Yates shuffle algorithm was implemented following the modern Durstenfeld variant, operating through backward iteration, generating random indices, and performing in-place element swapping. Algorithm validation was conducted through simulation calculations and chi-square goodness-of-fit statistical testing across ten thousand randomization trials. The application "Tebak Kosakata" successfully integrates the randomization algorithm with an accessible user interface, featuring multimodal content presentation, immediate positive feedback mechanisms, and cumulative scoring systems. Simulation calculations confirmed that each vocabulary item maintains an equal probability for occupying any position in the final sequence. Statistical validation yielded a chi-square value of 8.47 with nine degrees of freedom and a probability value of 0.487, confirming uniformly distributed randomization without detectable bias. The algorithm achieves optimal computational efficiency with linear time complexity and constant auxiliary space complexity. The randomization of question sequences and answer option positions effectively prevents pattern-based response strategies, encouraging authentic vocabulary learning rather than positional memorization. This study establishes that the Fisher-Yates shuffle algorithm constitutes an effective mechanism for implementing unbiased randomization in educational games for children with disabilities, bridging computational algorithm theory with special education pedagogy while providing a replicable methodological framework for future development.

Implementation of Tesseract OCR for Automated Payment Validation in E-Commerce

2026-05-05T14:09:50+00:00

The rapid expansion of e-commerce in Indonesia has resulted in a significant increase in digital transactions, necessitating expedited and precise payment verification. Administrators at the SweetJab hijab e-commerce platform must manually verify bank transfer receipts, a process that is time-consuming and susceptible to errors. This study utilises Optical Character Recognition (OCR) with the Tesseract engine as a supplementary approach for verifying transfer payments on the SweetJab website. The methodology encompasses image preprocessing (resizing to 200%, converting to greyscale, and enhancing contrast), employing Tesseract OCR with PSM 6 and an LSTM model for character recognition, and utilising regular expressions (regex) to extract structured transaction data. We employed Black Box Testing and Character Error Rate (CER) computations on 40 preliminary test samples and 40 post-implementation samples to assess the system. The initial test demonstrated an accuracy of 89.5%, which increased to 92.5% upon complete system integration. This study demonstrates that OCR is an effective method for extracting information from payment receipts, while maintaining security through a final manual verification by the administrator.

Investigating LOLBAS-Based Malware Using Hybrid Analysis: A Case Study of PowerShell-Driven Fileless Execution

2026-05-05T14:09:51+00:00

This study aims to identify and understand the technical characteristics of the malware output.exe, obtained from the MalwareBazaar repository, through a hybrid reverse engineering approach. This method combines static and dynamic analyses to provide a comprehensive understanding of the malware’s internal structure, execution behavior, and evasion techniques. Static analysis revealed the invocation of system functions such as CreateProcessW and RegSetValueExA, as well as the use of syscall to execute PowerShell commands directly, indicating the implementation of the LOLBAS (Living off the Land Binaries and Scripts) technique. Dynamic analysis using CAPE Sandbox confirmed the malware’s actual behavior, including process injection into legitimate processes such as svchost.exe, launching powershell.exe for data compression, and establishing network communication via Discord Webhook for data exfiltration. Integration of both analyses shows that output.exe functions as an information stealer with fileless execution and advanced persistence mechanisms. These findings demonstrate that the hybrid analysis approach is effective in identifying modern malware that leverages legitimate system components to evade traditional signature-based detection methods.

User Satisfaction level of Self-lending Machine using the End User Computing Satisfaction (EUCS) Model at the National Library of the Republic of Indonesia

2026-05-05T14:09:51+00:00

Libraries have undergone a significant transfomation from manual systems to automated services as part of advances in information technology. One innovation in circulation services is the implementation of self-lending machines to improve efficiency and user independece. This study aims to evaluate the utilization of self-lending machines at the National Library of the Republic on Indonesia using the End User Computing Satisfaction (EUCS) model, which includes ontent, accuracy, format, ease of use, and timeliness. This research employes a quantitative approach with a descriptive method. Data were collected through quastionnaires distributed to 86 respondents selected using purposive sampling. The results show that overall user satisfaction is categorized as ver satisfied. The content and accuracy dimensiones obtained the highest mean scores, followed by format and ease of use, while timeliness recorded the lowest score. These findings indicate that the system has been effectively implemented and is able to meet user needs, altough improvements are still required in terms of system responsiveness and usability. This study highlights the importance of continuous evaluation to enchance technology-based library services and improve user-oriented service quality.

Design and Implementation of a Shrimp Pond Monitoring Information System Using Internet of Things and Android Application

2026-05-05T14:09:52+00:00

Shrimp ponds are a place built to cultivate shrimp, with a tropical climate that makes Indonesia one of the largest shrimp producers, both saltwater, freshwater and brackish water shrimp. For shrimp farming business people, it is one of the profitable business opportunities because it is easy to maintain and produces fantastic profits. Maintenance for shrimp pond cultivation can be done by checking water quality using litmus paper, a digital pH meter or a chlorine test kit. However, this method is still considered less efficient because water is taken from the pond repeatedly over time, which wastes time and energy, in addition, changes in water quality in the pond can occur at any time, causing shrimp death and causing farmers to fail to harvest. In this study, water temperature, water pH, and water current sensors were used to monitor water quality. Then the data from the water quality detected by the sensor will be received by the Esp32 microcontroller and then send the data to firebase. Firebase plays a role in storing and sending data to Android so that it can be displayed on a smartphone. Poor data values will cause a notification to appear on the farmer's smartphone so that the farmer does not need to check the pond location repeatedly.