ejournal.nusamandiri.ac.id (STMIK Nusa Mandiri)
Not a member yet
1504 research outputs found
Sort by
The HYBRID CONTENT-BASED FILTERING AND CLASSIFICATION RNN WITH PARTICLE SWARM OPTIMIZATION FOR TOURISM RECOMMENDATION SYSTEM
No full textEconomic recovery in the tourism sector after the COVID-19 pandemic is one of the main focuses of the Indonesian government at the moment, especially in Bandung City. This research aims to develop a personalized tourist spot recommendation system, by addressing the gaps in the existing literature through the integration of Content-Based Filtering (CBF) and Simple Recurrent Neural Network (RNN) methods that aim to improve recommendation accuracy. This study uses a hybrid approach that combines Term Frequency - Inverse Document Frequency (TF-IDF) and word embedding with the Robustly Optimized BERT (RoBERTa) model to identify similarities between tourist destinations based on their content characteristics. Simple RNN is used to analyze user preference patterns over time, which is then further optimized using Particle Swarm Optimization (PSO). As a result, the Simple RNN model that has been optimized with PSO shows an increased accuracy of up to 94.37%, outperforming other optimizations such as Adam and SGD. This research makes a novel contribution by applying advanced machine learning techniques to improve personalization in travel recommendation systems
ANALYSIS STUDENT EMOTIONS AND MENTAL HEALTH ON CUMULATIVE GPA USING MACHINE LEARNING AND SMOTE
Get PDFThis research investigates the impact of emotions and mental health on students' cumulative grade point average (CGPA) using machine learning classification algorithms while addressing data imbalances with the Synthetic Minority Oversampling Technique (SMOTE). Emotional well-being and mental health are acknowledged as vital determinants of academic achievement. Data imbalance, particularly in mental health metrics such as anxiety and depression, frequently compromises forecast accuracy. This study improves the accuracy of CGPA prediction based on emotional and mental health factors by utilizing SMOTE in machine learning models such as logistic regression and random forest. A dataset including 226 university students, including academic records and self-reported mental health evaluations, was evaluated. The random forest model attained an accuracy of 87.63%, exceeding the logistic regression model's accuracy of 86.56%. These findings emphasize the significant role of emotions and mental health in academic outcomes and validate SMOTE’s efficacy in addressing class imbalance. This work offers a fresh technique in educational data mining by revealing the possibility for improved academic achievement forecasts based on psychological characteristics, helping to the development of targeted therapies for students experiencing emotional issues. Implications for educational policy emphasize the significance of mental health support systems in promoting academic achievement. Subsequent research should investigate supplementary psychological variables and comprehensible models to improve predictive accuracy and facilitate evidence-based policymaking
USABILITY EVALUATION OF MOBILE MULTI-FACTOR AUTHENTICATION BASED ON FACE AUTHENTICATION, GEOLOCATION AND QR CODE
Get PDFThe swift progress of information technology has led to the adoption of mobile-based multi-factor authentication (MFA) systems for attendance management, addressing inefficiencies, security issues, and errors inherent in traditional methods. By utilizing multiple layers of authentication—such as face recognition, geolocation, and QR code scanning—these systems significantly enhance security and reliability. This study evaluates the usability of a mobile MFA system, focusing on user-friendliness and learnability. Two iterations of the system were tested using cognitive walkthrough approaches, chosen for their effectiveness in simulating the experience of new users and identifying usability issues in system learnability. The initial version of the system utilized MobileFaceNet_v2, which had an input size of 112x112. This resulted in a false acceptance rate (FAR) of 0.26, a false rejection rate (FRR) of 0.2, and a half total error rate (HTER) of 0.23. Failures in face verifications and inadequate instructions led to significant user dissatisfaction. In the second iteration, improvements were made by providing better instructions during location and QR scan steps, adding a face capture confirmation screen, and increasing the input size of the face anti-spoof detection model to 224x224. This reduced the FAR to 0.11 but increased the FRR to 0.4, resulting in HTER to 0.25. While these updates improved security, usability issues such as ambiguous user feedback and inadequate instructions persisted. These results emphasize the need for an integrated approach that combines both technological improvements in authentication models and enhancements in UI design to create a more user-friendly experienc
ANALISIS STRATEGI BAURAN PEMASARAN PADA PERUSAHAAN PRODUKSI SUSU DAN MAKANAN DI ERA INDUSTRY 4.0
Get PDFThe Industrial Revolution 4.0 has driven a digital transformation that has brought significant changes across various sectors, including marketing. This study explores the marketing mix strategies employed by PT Cimory to address the challenges and opportunities of the Industry 4.0 era. With rapid technological advancements and the dynamic nature of consumer behavior, it is crucial for companies to adapt and innovate. This study aims to analyze how PT Cimory leverages the components of the marketing mix—product, price, place, and promotion—to enhance its market share and competitive advantage. The research adopts a qualitative design, utilizing secondary data from PT Cimory's annual reports, journal articles, and online media sources. The findings indicate that PT Cimory's strategic integration of technological advancements and innovative marketing approaches has significantly influenced consumer behavior and purchasing decisions, thereby strengthening its market position
MULTICLASS CLASSIFICATION FOR STUNTING PREDICTION USING DEEP NEURAL NETWORKS
Get PDFStunting is a chronic nutritional issue that hinders child growth and leads to serious long-term health and developmental impacts, particularly in developing countries. Therefore, early and accurate prediction of stunting is crucial for implementing effective interventions. This research aims to develop a multiclass classification model based on Deep Neural Networks (DNNs) to predict stunting status. The model is trained using a comprehensive dataset that encompasses various health variables related to stunting. The research process includes data collection, data preprocessing, dataset splitting, and training and evaluation of the DNNs model. The model can classify stunting status into four categories: stunted, severely stunted, normal, and tall. Further analysis is conducted to evaluate the influence of various parameters on the model's performance, including dataset splitting ratios (80:20 and 70:30) and learning rates (0.001, 0.0001, and 0.00001). The results show that a learning rate of 0.0001 yields the highest prediction accuracy, at 93.64% and 93.83% for the two data-splitting schemes. This indicates that this learning rate has achieved an optimal balance between convergence speed and the model's generalization capability. Additionally, the developed DNNs model can identify complex patterns hidden within the data without being affected by noise. These findings confirm that appropriate parameter selection, particularly the dataset splitting ratio and learning rate, can significantly enhance the DNNs model's ability to identify complex data patterns
EVALUASI USABILITY WEBSITE BUKALAPAK DAN TOKOPEDIA MENGGUNAKAN METODE USER EXPERIENCE QUESTIONNAIRE (UEQ)
Get PDFThis research aims to compare elements of usability and user experience on the e-commerce platforms Bukalapak and Tokopedia. Usability quality has important significance because if an application or website has optimal usability, users will be able to use it easily. Users who feel disappointed or have difficulty interacting with a website are likely to leave the platform and look for an alternative that is easier to use. Therefore, usability evaluation becomes crucial for e-commerce websites. Measurements were carried out using the User Experience Questionnaire (UEQ) method which involves several aspects, including attractiveness, perception, efficiency, dependability, stimulation and novelty. The results of evaluating user usage and experience on these two e-commerce websites, using the UEQ method, show that Bukalapak achieved an Attractiveness value of 0.80, Perspicuity of 0.83, Efficiency of 0.63, Dependability of 0.53, Stimulation of 0.74, and Novelty of 0.38. Meanwhile, Tokopedia achieved an Attractiveness score of 1.24, Perspicuity of 1.08, Efficiency of 1.20, Dependability of 1.19, Stimulation of 0.96, and Novelty of 0.70. Therefore, it can be concluded that Tokopedia shows better performance than Bukalapak in terms of usability and user experience on the e-commerce platform
PENERAPAN ALGORITMA CNN MENGGUNAKAN FRAMEWORK YOLO UNTUK DETEKSI OBJEK PRODUK DI PERUSAHAAN MANUFAKTUR
Get PDFComponent products used for manufacturing a machine in manufacturing companies have two types of products, type A and B. The problem that often occurs in the industry is product sorting errors due to the traditional sorting process, using human labor. The disadvantages are limited human labor so that fatigue can occur, causing errors in sorting products and losses for the company. Many studies discuss object detection, Industrial problems in the checking process can be approached with the help of this technology. Object detection works to analyze frames with the method of finding objects. There are methods in digital image processing, CNN algorithms which include methods in computer vision. The growing framework makes the CNN algorithm more powerful. YOLO includes a framework based on the CNN algorithm. YOLOv5 detects objects by taking into account the object's confidence value, the output of the detected object is a bounding box on the object. The problem in the industry in the checking process can be approached with the help of this technology. For this reason, this research aims to create a model for product object detection in manufacturing companies. The process carried out is data collection, image annotation, training, testing, evaluation. The images collected were 137 for training data and 34 for validation data totaling 171 image data. The results of the model using YOLOv5 with epoch 1000 get a precision value of 100%, recall 100% and mAP 99%, the product detection results get an average value of 100%
ABILITY CONVOLUTIONAL FEATURE EXTRACTION FOR CHILI LEAF DISEASE USING SUPPORT VECTOR MACHINE CLASSIFICATION
Get PDFChili plants are among the most commonly used food ingredients in various dishes in Indonesia. Leaves on chili plants are often affected by disease; if the disease is not treated immediately, it can damage the plant and cause crop failure. Early detection of chili plant diseases is important to reduce the risk of crop failure. The development of technology and the application of machine-learning algorithms can automatically monitor chili plants using a computer system. Using this algorithm, the system analyzes and identifies diseases that a camera can observe and record. In this study, the proposed method for feature extraction uses a convolutional neural network (CNN) algorithm with transfer learning using VGG19. For classification using SVM for training data, accuracy generated 95%, precision 95%, recall 95%, and F1-Score 95%, and testing data accuracy generated 90%, precision 89%, recall 90%, and F1-Score 89%, proving that the convolutional process with architecture VGG19 and SVM algorithm is acceptable for classification. In future research, other architectures or extraction fusions can be used to maximize the results
APPLICATION OF FUZZY LOGIC AND GENETIC ALGORITHM APPROACHES IN EVALUATION OF GAME DEVELOPMENT
Get PDFThe gaming industry is undergoing rapid evolution, presenting developers with intricate challenges in selecting compelling and successful game concepts. To tackle these challenges, decision support systems (DSS) play an increasingly crucial role in facilitating accurate decision-making. Despite their growing importance, the adoption of DSS within the gaming sector remains limited. Therefore, scientific research focused on developing DSS to evaluate optimal game concepts is essential to foster innovation in gaming industries. This study aims to construct a decision support system utilizing fuzzy logic and optimized with genetic algorithms to assess and identify game concepts with the highest potential for success in the market. Evaluation results highlight the system's effectiveness in recommending top-quality games like "Clash of Clans," "Honor of Kings," and "Genshin Impact," renowned for delivering exceptional gaming experiences and receiving high ratings. The system evaluation achieved an average Mean Squared Error (MSE) of 0.0246, indicating accurate prediction of game ratings with minimal error. The significance of this research extends beyond advancing decision support systems in gaming, opening avenues for further advancements in optimizing game evaluations and similar technologies across industries grappling with data-driven decision-making challenges
COMPARATIVE ANALYSIS OF EXPONENTIAL SMOOTHING MODELS FOR SALES PREDICTION AND SUPPLY MANAGEMENT IN E-COMMERCE
Get PDFIn the growing era of e-commerce, stock management is crucial. Problems arise in forecasting sales in order to achieve effective stock management. This research uses the time series analysis method by focusing on comparing the accuracy of three forecasting methods: Single Exponential Smoothing (SES), Double Exponential Smoothing (DES), and Triple Exponential Smoothing (TES/Holt-Winter). This research provides a solution by comparing the performance of the three methods based on the Mean Absolute Error (MAE) results and prediction graphs. The goal is to determine the most accurate forecasting method using the time series analysis method with several stages, namely data preprocessing, train/test split, modeling, and performance metrics measurement. based on the test results show MAE SES 1077, DES 96, and TES (Holt-Winter) 101. Although DES has a lower MAE, TES (Holt-Winter) provides better accuracy, especially through prediction graph analysis. Holt-Winter is recognized as the most effective method in forecasting future sales, reliable for proper stock management in the dynamic e-commerce industry. This approach is expected to improve efficiency and accuracy in enterprise stock management, support the growth of online businesses, and contribute to the literature and practice of stock management. The use of time series analysis methods, especially Holt-Winter, is considered an important strategic step to optimize sales prediction, positively impact stock management, and create a competitive advantage in a growing marke