Journal of Computer Networks, Architecture and High Performance Computing
Not a member yet
473 research outputs found
Sort by
Implementation of Machine Learning Models for Predicting Internet Service Provider Customer Churn
Full text linkThe telecommunications industry faces an extremely high level of competition, where the phenomenon of customer churn presents a significant challenge due to its impact on revenue decline and increased costs associated with acquiring new customers. This study aims to develop a churn prediction model using the Decision Tree algorithm and implement it in a web-based application to support customer retention strategies. The CRISP-DM methodology is employed, covering Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, and Deployment. Experimental results show that the Decision Tree algorithm demonstrates strong performance in identifying non-churn customers, with a precision of 0.82, a recall of 0.91, and an F1-score of 0.86. However, its performance on the churn class remains limited, with a precision of 0.63, a recall of 0.44, and an F1-score of 0.52, highlighting the importance of addressing imbalanced data distribution to preserve existing data. The model underwent Learning Curve and Validation Curve analysis. The Learning Curve indicates a relatively stable model with a small gap, suggesting good generalization. The Validation Curve reveals that optimal performance is achieved at a moderate tree depth, avoiding the risk of overfitting at greater depths. Nevertheless, the main advantage of the Decision Tree is its interpretability, which highlights significant factors such as contract type, subscription duration, and additional services. The integration of the model into a web-based application also provides practical benefits through rapid churn risk monitoring, supporting the company’s strategic decision-making
Implementation of a Desktop-Based Sales Information System Using the Waterfall Method
No full textIn today's digital era, information technology has become an integral part of various aspects of life. One of the many utilizations of information technology products is a desktop-based information system. The purpose of this research is to implement a desktop-based sales information system to solve the problem of data redundancy, calculation errors in sales transactions, and excessive use of time in the process of managing sales data in the Culinary business field because it still utilizes conventional methods. The sales information system developed in this study was built using the Visual Basic Net programming language. Visual basic net is an object-oriented programming language designed to build a variety of windows applications. The database used in the sales information system in this research is Microsoft Access. The collection methods used to obtain the required research data include observation, interviews, and literature studies. The system development method that the author uses to build a desktop-based sales information system in this research is the Waterfall method. The results of this study are in the form of a desktop-based sales information system equipped with various features to make it easier for users to systematically process sales data in the culinary business field. The desktop-based sales information system produced in this research allows faster and more accurate data management compared to conventional methods
The Implementation of Interactive Learning Media Utilizing Quizizz at SMAN 2 Rambah
Full text linkThe aim of this research is to find out how effective the use of Quizizz-based interactive learning media is by class XI Science students at SMAN 2 Rambah in the Informatics subject with computer network material. A quantitative method was used, with a questionnaire created by the school's Guidance and Counseling (BK) teacher to measure the level of student boredom before and after Quizizz. A Likert scale was used for data collection. Descriptive analysis, normality test, homogeneity of variance test, and difference test (paired sample t-test) were used to analyze the data obtained. All these techniques are used with the help of the SPSS program. The results show that Quizizz increases student participation in learning and significantly reduces student boredom. Before Quizizz was used, the average burnout score was 41.96. However, after use, this score dropped to 19.79. It was shown by the difference test that this difference was significant (p < 0.05). With this reduction in boredom levels, it is clear that Quizizz not only makes learning more interesting, but also makes learning fun and competitive. However, obstacles such as teachers' limited digital skills and difficulties in obtaining technology in remote areas still need to be overcome. According to this research, Quizizz can be a great interactive learning tool that gets students more engaged in learning, especially in high school
Comparison of Lexical and Semantic Approaches for Relevance Measurement in Quranic Verse Translation Retrieval
Full text linkThis research explores the effectiveness of lexical and semantic approaches for relevance measurement in Quranic verse translation retrieval, focusing on Indonesian translations. Quranic verses encompass complex linguistic structures and diverse contexts, making precise retrieval challenging. Two retrieval methods were evaluated: lexical similarity, which focuses on exact word matches, and semantic similarity, which captures contextual meaning using word embeddings. The study utilized a dataset of Indonesian Quranic translations, preprocessed to normalize and tokenize text, with experimental queries derived from thematic exegesis on social responsibility. Evaluation was performed using precision, recall, and F1-score on top-5, top-10, and top-15 retrieved results. The lexical approach achieved perfect precision (100%) but exhibited lower recall (46%-58%), as it failed to retrieve relevant verses lacking exact matches. Conversely, the semantic approach demonstrated higher recall (56%-59%) and F1-scores (73%-74%) by identifying verses with contextual relevance, even in the absence of lexical similarity. The results reveal that while the lexical approach ensures precise matches, it overlooks semantic richness. The semantic approach, although computationally intensive, achieves greater contextual understanding. These findings highlight the potential for hybrid retrieval systems combining both approaches to enhance accuracy and relevance in Quranic information retrieval, supporting scholarly research and user engagement with Quranic content
Optimization of Electric Power Flow Analysis Using the Gauss-Seidel Method in a Numerical Approach
Full text linkThe availability of electrical energy is a fundamental requirement in modern society, supporting both daily life and industrial activities. To ensure efficient and reliable energy distribution, power flow analysis is critical. This analysis is grounded in Kirchhoff's laws, which serve as the foundation for understanding electrical circuits. Kirchhoff's Current Law (KCL) states that "the sum of electric currents entering and leaving a branch point is zero," while Kirchhoff's Voltage Law (KVL) asserts that "the sum of electromotive forces and potential drops in a closed circuit must equal zero." These laws guide the formulation and solution of equations describing power flow in electrical networks. To manage the complexity of these systems, the Gauss-Seidel method has emerged as an effective iterative technique for solving large systems of linear equations. In the context of power flow analysis, it calculates busbar voltages, active and reactive power flows, and other parameters, refining the results through successive approximations until convergence is achieved. Python is widely recognized as an ideal platform for implementing the Gauss-Seidel method due to its syntactic simplicity, flexibility, and extensive computational libraries. By leveraging Python, engineers can streamline computations and enhance the accuracy and reliability of power flow analyses. This combination of mathematical rigor and computational power not only ensures precise results but also facilitates the efficient management of complex electrical systems in modern power grids
Research on Image Encryption Algorithm Based on Matrix Scrambling and Matrix Product Transformation
Full text linkThis study introduces an innovative image encryption algorithm that leverages multiple rounds of matrix scrambling and matrix product transformation. Each round of encryption integrates cross-scrambling operations within the image matrix and invertible matrix product transformations, thereby effectively disrupting pixel positions and values. By iteratively adjusting pixel positions and transforming pixel values, the algorithm significantly enhances the security and robustness of the encryption process. The experimental results demonstrate that the proposed algorithm exhibits excellent resistance to statistical analysis, differential attacks, and other potential threats, thereby ensuring high security and practical applicability
Comparison of Support Vector Machine, Random Forest and XGBoost for Sentiment Analysis on Indodax
Full text linkThe rapid growth of digital assets like Bitcoin and cryptocurrencies has increased the need for secure trading platforms such as Indodax. With the growing number of users, reviews on platforms like Google Play Store provide valuable insights into user experience and satisfaction. This research applies Machine Learning methods to classify user review sentiments by comparing three main algorithms Support Vector Machine (SVM), Random Forest, and Extreme Gradient Boosting (XGBoost). One of the main challenge in sentiment analysis is the presence of irrelevant or redundant features, which can reduce model accuracy and increase computational costs. The Feature Selection Chi-Square technique is used to filter the most influential features, enhancing model efficiency without losing critical information. Experimental results show that SVM delivers the best performance compared to Random Forest and XGBoost. Before applying Chi-Square, SVM achieved 91% accuracy, which increased to 94% after applying the feature selection technique. The number of features used was reduced from 52,312 to 2,000 without significant information loss. This combination of SVM and Feature Selection Chi-Square proves to be an efficient and accurate solution for analyzing user sentiment on crypto trading platforms like Indodax. This method is expected to improve the responsiveness of trading applications to user needs and serve as a foundation for further research in Machine Learning-based sentiment analysis
Analysis and Evaluation of SIDUN Mobile Application in UEQ-Based User Experience Perspective
Full text linkThis study analyzes and evaluates the user experience (UX) of SIDUN, a mobile-based Village Information System designed to manage community contributions digitally in Dusun Tegal Kori Kaja, Denpasar, Bali. The system aims to address limitations of the previous manual process by enabling digital interaction among villagers, pecalang, and administrative staff. The evaluation method applies the User Experience Questionnaire (UEQ), assessing six core dimensions: Attractiveness, Perspicuity, Efficiency, Dependability, Stimulation, and Novelty. A total of 30 active users participated in the study by completing the UEQ instrument. The results indicate that all six UX dimensions received positive scores, ranging from 1.58 to 2.00. The highest ratings were observed in Stimulation (2.00), Attractiveness (1.96), and Efficiency (1.92), reflecting high user engagement, visual appeal, and operational speed. Perspicuity and Novelty also showed strong performance, while Dependability, though positive, revealed opportunities for improvement in system reliability and consistency. Compared to the UEQ benchmark, all dimensions achieved “Excellent” ratings, placing them within the top 10% of evaluated applications. These findings affirm that SIDUN offers a satisfying user experience and supports effective community-level digital transformation. The study underscores the value of user-centered design and continuous UX assessment in enhancing public digital services in rural communities
Expert System Based on K-Nearest Neighbor for Oil Palm Fertilizer Application Optimization
Full text linkThis study aims to develop an expert system utilizing the K-Nearest Neighbor (KNN) algorithm to recommend suitable fertilizers for oil palm plants based on soil conditions, climate, and plant age. A quantitative approach was employed, involving literature review, data collection, model development, and evaluation. Data were obtained from PT. Nusantara Plantation IV Torgamba Plantation, including variables such as soil pH, dolomite, NPK, urea application, and crop yields. The KNN model was optimized with a K-value of 6 and evaluated using metrics including accuracy (63.63%), precision, recall, F1-score, Mean Absolute Error (MAE: 1995.38), and Mean Squared Error (MSE: 5,257,254.73). The system demonstrates the ability to provide fertilizer recommendations by identifying similarities in historical data, though further accuracy improvements are possible. The practical implications of this research include assisting farmers in optimizing fertilizer selection, enhancing productivity, and minimizing environmental impact. Future studies could explore the integration of additional variables or alternative algorithms such as Decision Tree or Naive Bayes to improve performance
Usability Evaluation of the SiCanTiK Website at SMKN 3 Kota Kediri Using the System Usability Scale and USE Questionnaire
No full textThe rapid advancement of information technology has driven the transformation of learning systems, especially in vocational education. SMKN 3 Kota Kediri has developed an e-learning platform called SiCantik (Sistem Informasi Canggih Terintegrasi dan Kolaboratif) to support the online learning process. However, despite its potential, the platform faces several usability issues such as difficulties in navigating features and accessing learning content, which may hinder learning outcomes. This study aims to analyze the usability level of the SiCantik website using the System Usability Scale (SUS) and the USE Questionnaire. A descriptive quantitative method was employed, and data were collected from 100 respondents through online questionnaires. Validity and reliability testing were conducted to ensure the accuracy of the instruments. The SUS results showed an average score of 61.8, which falls into the “Poor” category, indicating that the platform's usability is marginally acceptable. Meanwhile, the USE Questionnaire results produced an average usability percentage of 65.5%, which is categorized as “Feasible.” Among the four evaluated dimensions, Ease of Use had the highest score (68.5%), while Ease of Learning had the lowest (63.2%). These results imply that while users generally find the system usable, improvements are still needed, particularly in user guidance and system intuitiveness. This research provides valuable input for system developers and stakeholders to improve the user experience of educational platforms. Further research is recommended to evaluate the technical performance and long-term user engagement of the system