Sinkron : jurnal dan penelitian teknik informatika
Not a member yet
1261 research outputs found
Sort by
Performance Analysis of Random Forest Algorithm for Network Anomaly Detection using Feature Selection
No full textAs the volume and complexity of computer network traffic continue to increase, network administrators face a growing challenge in monitoring and discovering unusual activity. To keep the network safe and functioning, detecting anomalies is essential. Machine learning-based anomaly detection techniques have become increasingly popular in recent years. This is due to the fact that conventional anomaly detection methods make it difficult to detect unknown and complex attacks. This research aims to conduct a performance analysis of two feature selection methods using the random forest algorithm using the UNSW-NB15 dataset to determine which model is most effective in detecting network traffic anomalies. The models evaluated were random forest with the filter method and random forest with the wrapper method. A number of metrics used for model performance assessment are accuracy, F1-score, receiver operating characteristic curve, and precision-recall. Dataset collection, data pre-processing, feature selection, model construction, and evaluation are the main components of the research methodology. The research results show that the Random Forest approach with the Filter method has an accuracy of 0.8950, F1-score of 0.8333, ROC score of 0.8928, and a precision-recall value of 0.8347. Meanwhile, the approach using the Wrapper method obtained an accuracy of 0.9151, F1-score of 0.8510, ROC score of 0.9136, and a precision-recall value of 0.8637. This shows that the performance of Random Forest with the Wrapper method is superior in all assessment metrics. Random Forest with the Wrapper Method is the right choice of model for detecting network traffic anomalies because of its stable performance and ability to handle complex pattern
Prediction of Student Graduation Rates using the Artificial Neural Network Backpropagation Method
No full textThis student graduation rate research focuses on analyzing academic performance with the main aim of identifying and distinguishing between students who graduate on time and those who graduate late. The application of data mining techniques in this research uses the neural network method, which is expected to offer deeper insight into the factors that influence students' graduation times. In this study, the neural network method was used to classify graduation data from 150 students. The results of this analysis were very encouraging, with 149 students identified as graduating on time and one student graduating late. The level of accuracy achieved in this classification is 98%, which shows the effectiveness of the neural network method in processing and analyzing academic data. These results confirm that neural networks are a powerful and reliable tool for predictive tasks like this. The successful use of neural networks in this study also proves their potential in broader educational applications, particularly in optimizing educational and intervention strategies. By understanding the characteristics of students who graduate on time versus those who graduate late, educators and administrators can design more effective programs to support student success. This is important not only to improve graduation statistics, but also to improve the overall educational experience for students
Integrated MCDM-AHP and MABAC for Selection Head of Branch Offices
No full textLeadership changes are very urgent in maintaining organizational stability. A good relay can build significant strength in carrying out organizational operational activities, of course this must be done with good selection. The purpose of this writing is to provide a consistent picture of the selection of branch heads in carrying out business competition which is measured based on the competencies possessed by the selected employees. The barometer is determined based on eight criteria as an assessment that is declared objective by the leadership, namely critical thinking, communication, analyzing, creative and innovation, leadership, adaptation, cooperation, and public speaking. The method used will be implemented in an integrated manner from the two MCDM-AHP methods and the MABAC method. These two methods have similar applications to the selection process. MCDM-AHP is used to select eight criteria as determinants of weighting and the MABAC method is used to determine the ranking process assessment for integrated decision making. The results obtained based on the weighted matrices of the branch head office selection process were measurably obtained, namely that the first priority was held by A11 with a weight of 1,406. The results of the integrity of both methods can provide evidence of decision support for the branch head selection process consistently with optimal results. The ranking system can be regulated and utilized for the purposes of selecting leaders to be placed in other positions
Augmented Reality Learning Media Application In Computer Networking Courses
No full textIn computer network learning, there is still little use of media which has an impact on students' understanding of device material and computer network topology. Augmented Reality (AR) based learning media can answer these problems by providing dynamic visualization and interactive simulations. The research objective is that AR applications can be used to help visualize abstract concepts for understanding and structure of an object model. The development method used is MDLC (Multimedia Development Life Cycle) which consists of six stages, namely concept, design, material collecting, assembly, testing, and distribution. The results of the AR application research show that the value of the learning media application in terms of material is declared valid at 0.85 and in terms of design it is declared valid at 0.86. The AR application was also stated to be very practical, this can be seen from the responses of lecturers and students with the practicality of the learning media application being 87% as seen from ease, motivation, attractiveness, and usefulness. From the results of this research, the AR learning media application is very practical to apply to students, especially in computer networking courses.
 
Analyzing Public Sentiment Towards BSI Service Disruptions Through X: Naïve Bayes Algorithm
No full textDisruptions to banking services can negatively affect customer trust and happiness, thus affecting the bank's reputation in the eyes of the public. Analysis of sentiment expressed on social media is very important because it can provide a direct picture of individual perceptions and responses in real time. This research aims to analyze public sentiment towards disruptions in Bank Syariah Indonesia (BSI) services through social media using the Naive Bayes algorithm. Through this analysis, the research seeks to understand the pattern of public responses and perceptions of BSI disruptions and evaluate the performance of the Naive Bayes algorithm in classifying sentiment on related tweet data. The data used came from specific social media platforms, where sentiment analysis was conducted by categorizing the data into positive, negative, and neutral categories. The research findings show that the sentiment analysis of the community towards BSI service disruptions through X social media platforms shows a diverse pattern of responses and perceptions. This finding recorded 525 data points with negative sentiment, 325 data points with neutral sentiment, and 141 data points with positive sentiment. The research also compared the performance of the Naive Bayes algorithm with the Google Cloud Natural Language API, which showed an accuracy rate of 81.03%. This research provides valuable insights for Bank Syariah Indonesia in understanding public perception of BSI services on social media
Investigating the Impacts of A Simulation-Based Learning Model Using Simulation Virtual Laboratory on Engineering Students
No full textConsidering electrical engineering students at Universitas Negeri Medan as a case study, this research looks at how an SVL affected their grades. Integrating the TAM with the ABET Laboratory Learning Objectives, it provides a comprehensive framework for quality engineering and technology education. This research is the first of its kind to theoretically compare the two concepts in a VL context. This research examines the relationship between student performance in the classroom and the TAM's usability components as well as the ABET's learning objectives. The results from the surveys given to first-year Electrical Engineering students are analyzed using Structural Equation Modeling (SEM) and Partial Least Squares (PLS). Because it enhances student performance and satisfies their learning goals, the results demonstrate that utilizing simulation-based virtual laboratories (SVL) in engineering education offers substantial educational benefits
Comparison of Genetic Algorithm and Particle Swarm Optimization in Determining the Solution of Nonlinear System of Equations
No full textNonlinear systems of equations often appear in various fields of science and engineering, but their analytical solutions are difficult to find, so numerical methods are needed to solve them. Optimization algorithms are very effective in finding solutions to nonlinear systems of equations especially when traditional analytical and numerical methods are difficult to apply. Two popular optimization methods used for this purpose are Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). This study aims to compare the effectiveness of GA and PSO in finding solutions to nonlinear systems of equations. The criteria used for comparison include accuracy and speed of convergence. This research uses several examples of nonsmooth nonlinear systems of equations for experimentation and comparison. The results provide insight into when and how to effectively use these two algorithms to solve nonlinear systems of equations as well as their potential combination
Data Mining Clustering Analysis of Child Growth and Development Using the K-Means Method
No full textThis research aims to group children based on their growth and development characteristics. This method helps identify groups of children with normal growth and development, early signs of growth and development problems, and serious growth and development problems. The stages used in this research follow the Knowledge Discovery in Database (KDD) process, which consists of data selection, data pre-processing, data transformation, data mining, and evaluation or interpretation of results. By applying the K-Means method, this research aims to provide a clearer and more detailed picture of the distribution of children's growth and development problems and assist in decision making for more appropriate interventions. The K-Means method in data mining was used to group 102 sample data into three clusters based on children's growth and development characteristics. The results of this analysis show that 38 samples fall into Cluster 1 (C1), 36 samples into Cluster 2 (C2), and 28 samples into Cluster 3 (C3). Evaluation of clustering results is carried out using Box Plot and Scatter Plot. Box Plot shows a clear distribution of data for each cluster, ensuring that the data grouping corresponds to statistical evaluation. Cluster C1 is toddlers with normal growth and development. Cluster C2 shows early signs of growth and development problems. Cluster C3 indicates serious growth and development problem
Implementation of the C4.5 and Naive Bayes Algorithms to Predict Student Graduation
No full textThis research aims to determine student graduation using two data mining methods, namely the Naive Bayes Classifier and the C4.5 Algorithm. Research stages include data analysis, data pre-processing, model design in data mining, classification results, method evaluation, and evaluation results. This research uses student data consisting of training data and testing data to evaluate the performance of the two methods in predicting student graduation based on attributes such as attendance scores, behavior scores, Final Semester Examination (UAS) scores, and report card scores. The classification results show significant differences between the two methods. The Naive Bayes Classifier produces predictions that 37 students pass and 17 students do not pass, while the C4.5 Algorithm predicts that 30 students pass and 24 students do not pass. This difference in results indicates that there are differences in the approaches of the two methods to student graduation data, with the Naive Bayes Classifier tending to provide more positive predictions than the C4.5 Algorithm. Evaluation of the performance of the method shows that the Naive Bayes Classifier has an accuracy rate of 100%, which is a perfect result, while the C4.5 Algorithm has an accuracy rate of 89%. This significant difference in evaluation results confirms that the Naive Bayes Classifier is superior in classifying student graduation compared to the C4.5 Algorithm in the context of this research. These findings can help in making decisions regarding student graduation evaluations in the future
Comparison Of Machine Learning Algorithms On Stunting Detection For 'Centing' Mobile Application To Prevent Stunting
No full textStunting is a growth disorder caused by chronic undernutrition, with long-term impacts on child health and development. In Indonesia, the prevalence of stunting reached 31.8% in children under five years old in 2018, indicating an urgent need for effective interventions. In an effort to address this issue, we developed a mobile application called Centing (Cegah Stunting) that utilizes machine learning for early detection and prevention of stunting. In this study, we compare the performance of four machine learning algorithms Logistic Regression, Support Vector Machine (SVM) with Radial Basis Function (RBF) kernel, Convolutional Neural Network (CNN), and Multilayer Perceptron (MLP) in detecting children's nutritional status based on a dataset from Kaggle with 121 thousand data and four main features: age, gender, height, and nutritional status. The experimental results show that SVM with RBF kernel and CNN achieved the highest accuracy of 98%, while Logistic Regression and MLP achieved 76% and 97% accuracy respectively. SVM with RBF kernel was chosen as the best model due to its high accuracy and efficiency in computation time. These findings suggest that the Centing application, with the implementation of SVM RBF, has significant potential in early detection and prevention of stunting, and makes an important contribution to improving child health in Indonesia