Journal of Information Systems and Informatics (Journal-ISI)
Not a member yet
580 research outputs found
Sort by
Predictive Analytics on Shopee for Optimizing Product Demand Prediction through K-Means Clustering and KNN Algorithm Fusion
Get PDFThis study focuses on predictive analysis in the context of the Shopee market, aiming to optimize product demand forecasting through the combination of K-Means clustering and KNN algorithms. With the exponential growth of e-commerce platforms like Shopee, accurately predicting product demand is becoming increasingly important for inventory management and marketing strategies. In this research, we propose a novel approach that combines the strengths of K-Means clustering and the KNN algorithm to improve demand prediction accuracy. By leveraging K-Means clustering to group similar products into two clusters, namely “Low Interest” with 64 data points and “High Interest” with 25 data points, we then apply the KNN algorithm to predict demand within each cluster. The KNN algorithm produces two classifications: Low Sales and High Sales. Based on tests using the KNN algorithm with k values of 3, 5, and 7, it was demonstrated that the product “Soraya Bedsheet Cotton Gold Motif Dallas Ask Grey Tua” can be predicted to fall under “High Sales.” The sales prediction accuracy rate for Shopee marketplace products is 96%. The implications of these findings indicate that the combination of K-Means and KNN algorithms can improve the accuracy of product demand predictions and optimize inventory and marketing strategies
Decision Support System for Employee Performance Assessment Using Analytical Hierarchy Process and Simple Additive Weighting Methods
Get PDFEvaluation of employee performance is carried out by company leaders at company work meetings to determine the best employees. In the process of assessing the best employees, the relevant company still does not use a digital system and does not have a clear weighting system for the criteria used. This results in inaccurate assessment results. Based on the problems found during field observations, research was carried out with the aim of developing a computerized decision support system for employee performance assessment and a system that applies weighting to each criterion. The Analytical Hierarchy Process (AHP) and Simple Additive Weighting (SAW) methods are the methods used in developing this decision support system. The AHP method is used to determine the weight of each criterion, while the SAW method is used to determine the ranking of the best employees. The research results are in the form of a website-based decision support system complete with criteria and weights for each criterion as well as the final results of the performance assessment of the 5 data samples used. This system has been tested using the User Acceptance Test method and obtained a score of 78.31%
Implementation of the PIECES Framework as an Evaluation of Student Satisfaction Levels with the Use of STARS UKSW
Get PDFThis research aims to evaluate the level of student satisfaction with the use of the Academic Information System (STARS) at Satya Wacana Christian University (UKSW) using the PIECES (Performance, Information, Economy, Control, Efficiency, and Service) framework. The research method used was a survey with a structured questionnaire distributed to a sample of students from various study programs at UKSW. The collected data was analyzed using descriptive statistical techniques and the PIECES Framework to measure STARS performance from the user's perspective. The research results show that although STARS has provided good performance in several aspects, there are areas that require improvement to increase user satisfaction. The implication of this research is the importance of using a framework such as PIECES in evaluating information systems to thoroughly understand user needs and perceptions, as well as providing a strong basis for improvement and development of better systems in the future
Evaluation Virtual Assistant Chatbot Acceptance with an Unified Technology Acceptance and Use of Technology-Based Model
Get PDFThe virtual assistant chatbot in the MyTelkomsel application is a company's effort to achieve goals and improve business performance. Chatbot as a virtual assistant has advantages and disadvantages. Evaluating user acceptance of the Virtual Assistant Chatbot in the MyTelkomsel application is important to overcome so that users do not switch to using other products. One of the models used to evaluate user acceptance of information systems is the Unified Theory of Acceptance and Use of Technology (UTAUT). UTAUT is able to explain how the use of technology can be influenced by individual differences in use. The conclusion of this study is that the variables that influence the acceptance of the use of virtual assistant chatbots in the MyTelkomsel application consist of the greatest influence is shown in the effect of social influence on Behavioural intention with a value of 5,768. Then the second largest influence is the trust variable on behavioural intention with a value of 5,220. Meanwhile, the variable that has the smallest influence on behavioural intention is effort expectancy with a value of 5,134
Analysis of Customer Satisfaction with Marketing Services Using Fuzzy Logic
Get PDFIn sales companies, the acquisition of turnover every month is very influential on the assessment of sales quality. The problem faced by the company is a decrease in turnover, this may be caused by the performance of marketing services, therefore the purpose of this study is to evaluate the service performance of the marketing team at Jayaindo Abadi Makmur. To improve customer satisfaction, the team should consider components such as reliability, responsiveness, assurance, and empathy. To get the results, mamdani fuzzy logic is used with the stages of fuzzy set, implication function, rule composition, and affirmation (deffuzzy). The results showed that customer satisfaction with manual calculations amounted to 84.12, while the results with mamdani fuzzy logic using matlab software amounted to 81.3. The company's customer satisfaction is classified as very satisfied. Sales quality shows a decrease in turnover several times, but this is not caused by the marketing team. Recommendations for improvements that can be made include improving product management, pricing policies, and overcoming market competition. The data presented shows that the company's ability to manage products, pricing policies, and the competitive market atmosphere can contribute to higher levels of customer satisfaction
Enhancement Campus Office Supplies Requests Website Utilizing Rapid Application Development
Get PDFThe web-based office supplies requisition application is a significant step in modernizing office equipment procurement in various organizations, including universities. However, this website encounters problems due to immature planning and less effective implementation. The current website faces some problems with requesting office supplies on campus, where the current process lacks efficiency and transparency in the status of user requests. It often results in discrepancies between the registered stock of office supplies and the actual stock in the warehouse. Our research aims to improve this website using the Rapid Application Development methodology. We also include user feedback when designing this website. The result is a new web-based application that provides a much better user experience when requesting office supplies. This update is expected to increase the efficiency of office equipment request services, provide users with more transparent request status information, and ensure accurate stock availability
Comparative Analysis of Server-Based and Serverless Service Performance on Google Cloud Platform (GCP) (Case Study: Machine Learning Model Deployment)
Get PDFCloud infrastructure providers such as GCP provide various computing services to deploy applications such as machine learning models, namely server-based and serverless. However, the two services each have different characteristics and advantages so that this becomes a difficulty factor for users in choosing cloud services. This research was conducted to compare server-based and serverless services with the aim of knowing the best service resulting from the analysis of performance measurements, namely CPU and memory utilization, latency, pricing, and developer experiences. The application of machine learning models is carried out on Compute Engine and Vertex AI services and will be tested for performance through requests to endpoints 100 times using JMeter for 30 minutes. The findings show that Vertex AI performance is better than Compute Engine with CPU utilization of 0.10%, memory utilization of 0.94%, and latency of 17.34ms but the cost efficiency is owned by the Compute Engine
User Satisfaction and Application Usage of PA'KEPO: A UTAUT 2 Model Analysis
Get PDFThe PA'KEPO (Payo jadi Keluargo Polisi) application is a mobile Android application owned by Polda Sumsel for the recruitment process of police members. This study aims to evaluate user satisfaction and the use of the PA'KEPO application by applying the Unified Theory of Acceptance and Use of Technology (UTAUT) 2 model with the addition of the Perceived Trust variable, which represents users' trust in the security and reliability of the application, enhancing users' intention to use the application and impacting actual usage behavior. The analysis of the relationships between variables employs the Structural Equation Modeling (SEM) approach to test the complex relationships between latent variables, allowing for the analysis of data with diverse scales. The research results indicate that the majority of respondents experienced a high level of satisfaction with the PA'KEPO application. The most influential variables—Effort Expectancy, Perceived Trust, Hedonic Motivation, Habit, and Behavioral Intention—significantly affect Use Behavior. Based on these findings, it is recommended that Polda Sumsel continue to encourage the use of the PA'KEPO application by optimizing the variables that have not yet shown significant effects. Improvement recommendations include evaluating and enhancing the application's functionality in line with the daily needs of its users
Development of a Real-Time Traffic Density Detection Website Using YOLOv8-Based Digital Image Processing with OpenCV
Get PDFThis study introduces a real-time traffic density monitoring system utilizing YOLOv8-based digital image processing to improve traffic management efficiency. By leveraging YOLOv8’s enhanced speed and precision, the system detects and classifies five types of vehicles and displays traffic data through a web interface developed with OpenCV and Flask. Key implementation features include real-time video streaming and accurate detection metrics, with the system achieving 96% Precision, 84% Recall, and an F1 Score of 90% during field testing in Bogor. This indicates the system’s potential for minimizing manual traffic monitoring and aiding traffic authorities in making data-driven decisions. The research also discusses the system’s integration into urban traffic management and its scalability for diverse environments
Dynamic Segmentation Analysis for Expedition Services: Integrating K-Means and Decision Tree
Get PDFTechnological developments have an impact on increasing the level of competition between companies in acquiring and retaining customers. With this competition, companies must maximise efforts to reach consumers and understand customer service needs so that the business can continue to survive and experience development. In this effort, a segmentation analysis was carried out on marketplace accounts and expedition services commonly used by consumers to make transactions. The first step is to correct the dataset obtained to avoid errors in the final results. Next, data processing was done using rapidminer with the k-means clustering and decision tree methods. The research results show that k-means clustering achieved the lowest Davies Bouldin Index (DBI) accuracy, namely -0.943 in cluster_8. In the results of research using the decision tree method, accuracy results were obtained at 49.83%. The results obtained with the decision tree method cannot be said to be good because the results are below the 50% value; however, the decision tree method shows that a good cluster is cluster_7. In this case, better accuracy values can be achieved by using the k-means clustering method. This research can illustrate the importance of utilizing the k-means and decision tree algorithms in classifying sales data as a tool for optimizing marketing and service efforts