Journal of Information Systems and Informatics (Journal-ISI)
Not a member yet
580 research outputs found
Sort by
A Qualitative Study of Researchers Perspective on the Use and Risks of Open Government Data
Full text linkOpen government data has the potential to improve transparency, accountability, public participation, business innovation, and research quality. However, this openness also poses various opportunities for losses or even risks, especially related to low data quality, personal data security issues, data translation errors, and misuse of information. This study aims to review the potential risks of data openness on government data portals from the perspective of researchers as one of the important actors who use data. Using qualitative method with structured interviews, this study involved five potential researchers who actively used open data between May and August 2023. The results of the interviews showed that high data quality, such as accuracy, completeness, and currency, can increase researchers' trust in the data. At the same time, obstacles in accessibility and bureaucracy or data administration requirements can slow down the research process or stages. Security and privacy issues are also important parameters, with strict security policies and good audit processes can reduce the risk of data misuse. Data openness and transparency play a major role in increasing the use of data for public policy and evidence-based research. In addition, data standardization is essential to ensure the efficiency of data use by researchers. This study concludes that to optimize the benefits of data openness, there needs to be proper and measurable management in order to consider data quality, accessibility, security, and standardization
Real-Time Sign Language Recognition and Translation in Humanoid Robots Using Transformer-Based Model with a Knowledge Graph
Full text linkFor millions of deaf-mute individuals, sign language is the only means of communication; this creates barriers in daily interactions with non-signers, leading to the exclusion of these individuals in many areas of daily life. To address this, we propose a real-time sign language translation system using a Transformer model enhanced with a knowledge graph, designed for Human-Robot Interaction (HRI) with NAO robots. Our system bridges the communication gap by translating gestures into natural language (text). We used the RWTH-PHOENIX-Weather 2014T dataset for initial training, achieving a BLEU score of 29.1 and a Word Error Rate (WER) of 18.2% surpassing the baseline model. Due to the domain shift between human gestures and NAO robot gestures, we created a NAO-specific dataset and fine-tuned the model using transfer learning to accommodate an adapted environment and kinematic constraints that do not match the environment in which the robot was deployed. This reduced the WER to 17.6% and increased the BLEU score to 29.9. We tested our model’s capability with dynamic and practical HRI scenarios through comparative experiments in Webots. Integrating a knowledge graph into our model improved contextual disambiguation, significantly enhancing translation accuracy for gestures that weren't clear. Through effectively translating gestures into natural language, our system demonstrates strong potential for practical robotic applications that promote social accessibility
A Model for Digitization Success in Ugandan TVETs: Evaluation Through Structured Walkthroughs and Simulation
Full text linkThis study proposes an information systems model to enhance the success of digitization projects in Ugandan Technical and Vocational Education and Training (TVET) institutions. The research was based on agency theory, with additional insights drawn from the DeLone and McLean Information Systems Success Model and the Dynamic Capabilities Framework. The model was developed based on key constructs such as Communication, Task Programmability, Goal Conflict, Shirking, and Process Quality. To evaluate its effectiveness, a structured walkthrough was conducted using a prototype simulator (SimPro), where expert evaluators assessed its usability, completeness, and performance. Results indicate that 96% of experts rated the model as highly usable, while 92% agreed that it accurately represents key digitization principles. The model’s usability significantly influenced expert recommendations for adoption (Spearman’s rho = 0.457, p = 0.001). Based on expert feedback, refinements were made to enhance stakeholder engagement, accountability tracking, and task efficiency. These findings suggest that the model has strong potential to improve digitization success rates by enhancing stakeholder engagement, accountability tracking, and task efficiency. Expert evaluators confirmed that these factors are critical to successful digitization in TVETs, indicating that structured implementation of this model could lead to more effective digitization outcomes. However, further empirical validation through real-world implementation is recommended to measure long-term impact
Advanced Techniques for Anomaly Detection in Blockchain: Leveraging Clustering and Machine Learning
Full text linkBlockchain technology has revolutionized data security and transaction transparency across various industries. However, the increasing complexity of blockchain networks has led to anomalies that require further investigation. This study aims to analyze anomalies in blockchain systems using machine learning approaches. Various anomaly detection techniques, including supervised and unsupervised methods, are evaluated for their effectiveness in identifying irregularities. The results indicate that machine learning models can detect anomalies with high accuracy, providing insights into potential threats and system vulnerabilities. The findings of this research contribute to improving blockchain security and developing more robust monitoring systems
Forensic Investigation of Drug and Food Crimes in Digital Marketplace
Full text linkThis research holds great significance as it is anticipated to safeguard customers from harmful frug and food products in cyberspace and to enforce the law by offering solid evidence to bring criminals to justice. Furthermore, this research helps to better understand how crimes that take place in the marketplace are committed, which enables the implementation of more effective preventive measures. Besides, in order to combat cybercrime, the findings of this study may serve as the foundation for the creation of more effective digital forensic investigation techniques. In order to perform digital forensic investigations of drug and food offenses using the marketplace, this study aims to develop an efficient and successful model or implementation guideline. This seeks to methodically direct the inquiry process while adhering to relevant norms. The objective of this research endeavor is to provide a model or practical guideline that is both effective and efficient for using the marketplace to undertake digital forensic investigations of drug and food crimes. This seeks to methodically direct the research process while adhering to relevant criteria. The following stages make up the Design Science Research (DSR) method of research: The issue in this project is "How to conduct a digital forensic investigation for Drug and Food crimes using the marketplace so that it can be used as evidence in court?" Then, in order to adopt answers from related research and make adjustments linked to research difficulties, a literature review is conducted to locate prior research. A model or implementation guideline for performing digital forensic investigations of the marketplace is the type of solution or artifact anticipated in this project. The next phase is solution design, which involves using an existing forensic investigation framework to create an artifact design. Following every step of the framework, case study experiments are then conducted to test the artifact design. The examination of the artifact design by both experts and consumers is the last phase
Predicting Student Loyalty in Higher Education Using Machine Learning: A Random Forest Approach
Full text linkStudent loyalty is a crucial factor supporting the sustainability of higher education institutions. The aim of this study is to predict student loyalty using a machine learning approach, specifically the random forest algorithm. The data for this research were collected through a questionnaire that included variables such as service quality, emotional attachment, brand satisfaction, brand trust, and socio-economic conditions, distributed to 107 students in Palembang. The resulting dataset was processed through preprocessing, model training, and performance evaluation, employing metrics such as accuracy, precision, recall, and F1-score. The analysis using the random forest algorithm achieved an accuracy of 90.9%. These findings are expected to provide valuable insights for higher education institutions in developing more effective strategies to enhance student loyalty
A Comparative Study of Drug Prediction Models using KNN, SVM, and Random Forest
Full text linkAccurate drug classification is essential in medical decision-making to ensure patients receive appropriate prescriptions based on their physiological and biochemical characteristics. This study compares the performance of K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Random Forest models in predicting drug prescriptions using patient attributes such as age, sex, blood pressure, cholesterol level, and sodium-to-potassium ratio. The dataset, obtained from Kaggle, was preprocessed and split into training and testing sets to evaluate model performance using accuracy as the primary metric. The results indicate that Random Forest outperformed KNN and SVM, achieving a perfect test accuracy of 100%, demonstrating superior generalization and robustness. SVM also performed well, with a test accuracy of 97.50%, while KNN achieved the lowest accuracy of 70%, indicating its limitations in handling complex feature interactions. These findings highlight the effectiveness of ensemble learning methods in medical classification tasks, suggesting that Random Forest is the most suitable model for drug prediction. Furthermore, the potential applications of these findings in clinical settings could enhance treatment outcomes and patient care. Future research should explore feature engineering techniques, larger datasets, and additional machine learning approaches to enhance predictive accuracy and applicability in real-world healthcare settings
Clustering of High School Students Academic Scores Using K-Means Algorithm
Full text linkThe clustering of student subject scores in senior high school is conducted using the K-Means Clustering algorithm. The issue addressed in this study is how to optimally group students based on their academic scores to help schools understand the distribution of student abilities. This clustering is essential as a foundation for evaluating and improving the learning system. The research methodology includes data collection and preprocessing, determining the optimal number of clusters using the Davies-Bouldin Index (DBI), and applying the K-Means Clustering algorithm. The analysis results indicate that the optimal number of clusters is three, with an average DBI value of 1.226. Cluster 0 is categorized as "very good" (46 students), Cluster 1 as "good" (70 students), and Cluster 2 as "less good" (51 students).The clustering results can be utilized for more targeted learning interventions and curriculum adjustments. Schools can implement remedial programs or additional classes for students in the "less good" cluster to improve their academic performance. Meanwhile, students in the "very good" cluster can be provided with advanced learning materials or opportunities to participate in academic competitions. Additionally, clustering outcomes provide valuable insights for refining teaching strategies, allocating resources more effectively, and personalizing learning approaches to suit each student's needs. Furthermore, these clustering results support academic decision-making by enabling educators and administrators to identify student performance trends and address potential learning gaps. This data-driven approach helps schools enhance overall educational quality by adapting teaching methods and policies based on empirical findings
Innovating Cybersecurity in Tanzanian Academia: A Mobile Tool for Combatting Social Engineering Threats
Full text linkSocial engineering attacks, including phishing, smishing, and vishing, pose significant threats to higher learning institutions, especially in regions with limited cybersecurity awareness and weak incident reporting mechanisms. This study introduces a novel mobile tool that combines real-time threat detection, streamlined reporting, and personalized training to address these vulnerabilities. Using a mixed-methods approach, we gathered survey data from 395 participants, conducted interviews with 10 IT professionals, and ran a pilot test with 20 users. The proposed tool provides instant scanning of emails/SMS for social engineering content and instant incident reporting alongside interactive, bilingual (English/Swahili) training modules. Results show a substantial improvement in user awareness, 85% of users reported a better understanding of social engineering threats after using the app, and high user satisfaction, with 90% expressing approval of the intuitive interface. The integration of real-time threat analysis and immediate reporting with tailored education distinguishes our tool from existing solutions. We discuss how bilingual support broadened engagement and how personalized learning paths reinforced retention of security best practices. Our findings demonstrate that a mobile-based, user-centric approach can significantly bolster cybersecurity awareness and incident response in academic environments. Future work will integrate machine learning for enhanced threat detection and voice-guided features for accessibility, aiming to continuously adapt to evolving attack strategies. This research provides insights for policymakers on incorporating such tools into broader institutional cybersecurity strategies
Developing a Cloud-Native Internship Management Platform: Enhanced Efficiency and Integration through Object-Oriented Architecture
Full text linkThe Internship Program Fieldwork Practice plays a crucial role in bridging academic learning and industry experience. However, traditional internship management faces challenges such as inefficient supervision, inconsistent attendance tracking, and lack of standardized performance evaluations. This study proposes a cloud-native internship management platform utilizing Object-Oriented Design (OOD) to enhance efficiency and accuracy in administrative processes. Developed using the Rapid Application Development (RAD) methodology, the system provides real-time monitoring, automated attendance tracking, and centralized performance assessment. User acceptance testing involving 35 participants, including students, lecturers, and industry supervisors, revealed significant improvements in administrative efficiency, student engagement, and data security. The platform ensures scalability, role-based access control, and secure data encryption. Findings highlight the need for standardized, technology-driven internship management solutions. Future research should explore AI-driven analytics and machine learning for optimizing internship experiences