Journal of Information Systems and Informatics (Journal-ISI)
Not a member yet
580 research outputs found
Sort by
Harnessing SVM for Sentiment Analysis: Insights from Gojek's Instagram Engagement
Full text linkThe development of digital technology has changed the transportation industry, including online services such as Gojek. Understanding customer sentiment is key in improving user experience and designing more effective business strategies. This research analyzes Gojek user sentiment on Instagram using Support Vector Machine (SVM). Data is obtained through web scraping, then processed through text cleaning, tokenization, common word removal, and stemming. Features were extracted using Term Frequency-Inverse Document Frequency (TF-IDF) before being classified with SVM. The results showed that the SVM model achieved 70.82% accuracy in classifying user sentiment. Most positive comments highlight the convenience and efficiency of the service, while negative comments are more related to high tariffs, application constraints, and less responsive customer service. These findings provide insights for Gojek to improve marketing strategies, optimize customer service, and adjust fare policies based on user feedback. In addition, this analysis can help in predicting real-time customer satisfaction trends through sentiment monitoring on social media. As a development step, this research recommends further exploration with deep learning and Aspect-Based Sentiment Analysis (ABSA) to improve accuracy and understand the service aspects that have the most influence on customer satisfaction
Enhancing User Satisfaction and Loyalty in MSMEs: The Role of Accounting Information Systems
Full text linkThis study examines the impact of Accounting Information System (AIS) attributes on User Satisfaction, Decision-Making, and Loyalty among Micro, Small, and Medium Enterprises (MSMEs). The research evaluates how Content, System Quality, Information Quality, and User Characteristics influence satisfaction and decision-making, ultimately shaping user loyalty. A quantitative approach utilizing Partial Least Squares Structural Equation Modeling (PLS-SEM) was employed to analyze responses from 62 MSME operators in East Lombok, Indonesia. The findings indicate that Content and System Quality significantly enhance User Satisfaction and Decision-Making, which in turn mediate their effects on User Loyalty. Contrary to conventional expectations, Information Quality has a minimal impact, suggesting that MSMEs prioritize system usability and functionality over informational attributes. The study reinforces the critical mediating roles of satisfaction and decision-making, highlighting how system attributes influence behavioral outcomes. In practical terms, AIS should be designed with simplicity, reliability, and relevance to MSME needs, ensuring ease of adoption and operational efficiency. Policymakers are encouraged to promote digital literacy and provide affordable AIS solutions to accelerate adoption among MSMEs. Additionally, the study suggests that future research explore broader cultural and organizational dynamics affecting AIS adoption and employ mixed-method approaches for deeper insights into user behavior
Heap Optimization in A* Pathfinding for Horror Games
Full text linkThis paper examines the implementation of the A* pathfinding algorithm with binary heap optimization in a horror game environment. The horror genre in gaming uniquely engages players by placing them at the center of fear-driven experiences, where intelligent and unpredictable enemy behavior is critical for immersion. To achieve this, adaptive AI—specifically for apparitions or monsters—is controlled using A*, an algorithm renowned for its efficiency in determining the shortest path. Heap optimization is introduced to enhance A* performance by reducing the time required to identify the lowest-cost node in the Open List. Experimental results from a Unity-based prototype demonstrate that the optimized A* achieves an average pathfinding time of 1.6 ms, compared to 3.16 ms without optimization—representing a 49.37% improvement. This speed increase allows for faster and more responsive enemy behavior, resulting in heightened difficulty and more dynamic, fear-inducing gameplay. The findings highlight the potential of algorithmic optimization to significantly enhance both technical performance and player immersion in horror game design
Exploring Internet Radio’s Impact on Dispersed Communities in Ghana
Full text linkDue to the constant challenges that migrants face in their host country, which sometimes lead to alienation, they strive to find solace in internet radio connecting them to happenings in their home countries as well as the world. Satisfaction with information systems (IS) has long been a topic of research in the discipline. It has primarily been employed as a stand-in for IS success. Researchers have utilized DeLone and McLean’s 2003 extensive model of factors to evaluate IS performance and the interrelationships between the variables. According to their approach, the success of IS is largely dependent on user pleasure. They also suggested that the main antecedents of user happiness are system, information, and service success. Drawing clues from the major components of Delone and McClean’s IS success model, the article explores factors that influence the use of internet radio as an information system. The literature review reveals some attribute levels (accessibility, empathy, trustworthiness, etc.) that are confirmed and amended by the empirical study. The study uses a qualitative approach using interviews to collect data that examines the impact of internet radio on people living in dispersed communities. The results prompt key attributes of IS success, which are used to chart the impact of internet radio. The findings reveal that service success attributes (accessibility, empathy, etc.), data success (understandability and relevancy, etc.), and technology success attributes (availability, ease of use, etc.) impact the use of internet radio. Using the attributes identified in the literature review, as confirmed by the empirical study, as well as three additional constructs that emerged during the semi-structured interviews, a framework is developed to determine the impact of internet radio on dispersed communities. The research presents a novel comprehension of the impact of internet radio by applying and extending multi-attributes from the Delone and McClean IS success model
Prioritizing Higher Education Facilities Using TOPSIS Based on Student Preferences
Full text linkThis research aims to prioritize campus facilities for development based on student preferences using the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method. Recognizing the critical role of facilities in enhancing student success and retention, this study evaluates key criteria such as needs, comfort, current conditions, accessibility, and frequency of use. Data were collected through a random sampling survey involving 98 active students, determined using Slovin's formula with a 10% margin of error. The analysis identifies WiFi as the top priority for improvement, followed by toilets and lifts. This research highlights how TOPSIS has been applied effectively in decision-making processes within education and facility management, offering a structured approach for optimizing resource allocation
Evaluating the Efficacy of AI Tools in Systematic Literature Reviews: A Comprehensive Analysis
Full text linkArtificial Intelligence (AI) tools can revolutionize literature review practices by transforming the research landscape towards more efficient and reliable review processes. While conducting literature can be challenging and time-consuming, there is a plethora of AI powered tools which uncover potential solutions to the challenge. AI tools may reduce the time spent on repetitive tasks, allowing scholars to focus more on critical analysis and interpretation. Due to the rising abundance of AI tools, it is difficult to decide which AI tools are best for individual research problems or projects. While concerns exist around the ethical and quality consequences of using AI. The study aims to explore the usage of AI tools on the systematics literature review process, specifically focusing on their effectiveness in various stages and ethical concerns. IEEE and MDPI Journal papers from 2020 to 2024 were reviewed using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. RobotReviewer, Covidence and EPPI-Reviewer are AI tools commonly used. These AI tools are designed to support different aspects of the systematic literature review process by offering capabilities such as problem formulation, literature search, inclusion screening and quality assessment. AI tools demonstrate improved effectiveness of literature searches, screening processes and data extraction. Language and content presentation, incorrect citation and plagiarism, grammar and spelling errors may be ren when utilizing AI. Concerns related to data quality, biases, and the need for human oversight were identified
A Data-Driven Framework for Optimizing Propranolol Dosage Using Support Vector Regression and Reinforcement Learning
Full text linkThe accurate prediction and adjustment of drug dosages requires precision to maximize therapeutic benefits while minimizing harm. This research attempts to model a hybrid machine learning framework combining Support Vector Regression (SVR) and Reinforcement Learning (RL) for individualized Propranolol dosage optimization using patient-specific clinical, enzymatic, and lifestyle data. A retrospective dataset comprising patient file, lifestyle indicators, and enzyme profile was used to train an SVR model for initial dosage prediction. Reinforcement Learning was subsequently applied to refine predictions through simulated feedback loops. Model performance was assessed using Mean Squared Error (MSE), R-squared (R²), and F1-score. Statistical comparisons between SVR predictions, RL-refined dosages, and physician-prescribed doses were performed using paired t-tests and one-way ANOVA. The SVR model achieved high predictive accuracy (MSE = 0.3554; R² = 0.9835), indicating its suitability for dosage estimation. The RL-refined model demonstrated a slight decrease in accuracy (MSE = 0.9928; R² = 0.9539). Statistical tests showed no significant improvement with RL (paired t-test: t = -1.1132, p = 0.2672; ANOVA: F = 0.0165, p = 0.9836). Mean predicted dosages across SVR, RL, and physician prescriptions were closely aligned (24.85 mg, 24.83 mg, and 24.93 mg, respectively). This study demonstrates that even standalone SVR may yield Propranolol dosage estimates with high accuracy, highlighting its prospective usefulness in clinical settings as a direct yet reliable tool for use in customized healthcare. While RL does offer some level of flexibility, the statistical value of improvements made was negligible, making RL beneficial but not necessarily critical. The proposed model shows that AI systems can aid in formulating evidence-based clinical judgments for dosing medications
A Systematic Literature Review on Machine Learning Algorithms for the Detection of Social Media Fake News in Africa
Full text linkFake news has been around in history before social media emerged. Social media platforms enable the creation, processing, and sharing of various kinds of content and information on the Internet. While the mediums of information and content shared across social media platforms are hard for users to authenticate, if users are tracking fake information or fake content, it can harm individuals, society, or the world. Fake news is increasingly becoming a worrisome issue, especially in Africa, because it's difficult to identify and stop the distribution of fake news. Due to languages and diversity, it is difficult for humans to understand and subsequently identify fake news on social media platforms, so high-level technological strategies, such as machine learning (ML), would be able to tell if the content is false material. As such, this study sought to identify effective ML classifiers to detect fake news on social media platforms, and the systematic literature review followed the PRISMA standard. The study identified 14 effective ML classifiers to manage fake news on social media platforms, including Random Forest, Naive Bayes, and others. Four research questions guided the study focused on the effectiveness of the classifiers, their applicability for detecting different forms of false news, the features of the dataset size and features, and the metrics that were created to assess the metrics. A conceptual framework known as the Information Behavioral Driven Social Cognitive Model (IBDSCM) was proposed in a bid to affect the fake news detection on social media platforms. Overall, this study establishes a contribution to understanding the ML algorithms for detecting false news in Africa and allows for a conceptual base for future studies
Technology Acceptance Model TAM using Partial Least Squares Structural Equation Modeling PLS- SEM
No full textThe rapid advancement of digital technologies necessitates a deeper focus on user acceptance and satisfaction, particularly within the framework of the Technology Acceptance Model (TAM), analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). This systematic literature review examines 36 articles published between 2020 and 2025, revealing that factors such as trust, system quality, perceived enjoyment, service quality, and technological self-efficacy significantly influence user satisfaction. These external variables enhance the explanatory power of TAM, providing a richer understanding of user interactions with digital platforms such as e-commerce, e-learning, and mobile banking. PLS-SEM's ability to manage model complexity, non-normal data distributions, and interrelated constructs further validates its suitability for this research. The findings suggest that integrating these external factors improves both the theoretical and practical aspects of TAM in the context of technology adoption. Future research could explore additional industry-specific applications for emerging technologies
Decision Support System for Job Applicant Recommendation Using ROC and ORESTE Methods
Full text linkThis study developed a Decision Support System (DSS) to assist the employee selection process for an agricultural company located in North Sumatra by utilizing a combination of the Rank Order Centroid (ROC) and ORESTE methods. The system was designed to address the limitations of the manual selection process, which had been time-consuming and inefficient. The ROC method was used to objectively determine the weights of each selection criterion, while the ORESTE method was applied to rank candidates based on their closeness to the company’s ideal profile. The study evaluated ten candidates based on key aspects such as educational background, competencies, work experience, and completeness of documents. The testing results demonstrated that the system was capable of producing accurate rankings consistent with manual calculations and was able to reduce the selection time from approximately two months to just a few minutes. The implementation of this system improved the objectivity and efficiency of the selection process while minimizing the risk of subjectivity in recruitment decision-making