JURNAL INFOTEL

Recursive Feature Elimination Optimization Using Shapley Additive Explanations in Software Defect Prediction with LightGBM Classification

Hartati Hartati — 2025-02-28

Software defect refers to issues where the software does not function properly. The mistakes in the software development process are the reasons for software defects. Software defect prediction is performed to ensure the software is defect-free. Machine learning classification is used to classify defects in software. To improve the classification model, it is necessary to select the best features from the dataset. Recursive Feature Elimination (RFE) is a feature selection method. Shapley Additive Explanations (SHAP) is a method that can optimize feature selection algorithms to produce better results. In this research, the popular boosting algorithm LightGBM will be selected as a classifier to predict software defects. Meanwhile, RFE-SHAP will be used for feature selection to identify the best subset of features. The results and discussion show that RFE-SHAP feature selection slightly outperforms RFE, with average AUC values of 0.864 and 0.858, respectively. Moreover, RFE-SHAP produces more significant results in feature selection compared to RFE. The RFE feature selection T-Test results are P_value = 0.039 < Î± = 0.05 and t_coun_t = 3.011 > t_table = 2.776. On the contrary, the RFE-SHAP feature selection T-Test results are P_value = 0.000 < Î± = 0.05 and t_coun_t = 11.91 > t_table = 2.776.

Object Detection of Hospital Assets Using Computer Vision with Generative Adversarial Networks Method

Sinung Suakanto — 2025-04-08

Hospital asset monitoring systems encounter significant challenges in managing partially occluded medical equipment, which affects inventory management and operational efficiency. Conventional object detection methods have shown limitations in accurately detecting occluded medical equipment, potentially leading to asset management inefficiencies. This study presents an integrated framework that combines Generative Adversarial Networks (GAN) inpainting with YOLOv8 to improve the detection accuracy of partially occluded medical equipment. The proposed system was evaluated using three distinct training configurations of 500, 750, and 1000 epochs on a comprehensive medical equipment dataset. The experimental results indicate that the 1000-epoch GAN model demonstrated superior reconstruction performance, achieving a Peak Signal-to-Noise Ratio (PSNR) of 39.68 dB, Structural Similarity Index Measure (SSIM) of 0.9910, and Mean Squared Error (MSE) of 7.0030. Furthermore, the integrated YOLOv8-GAN framework maintained robust detection performance with an F1-score of 0.933, comparable to the 0.938 achieved with unoccluded original images. The detection confidence scores exhibited improvement at higher epochs, ranging from 0.824 to 0.861, suggesting enhanced performance with extended training duration. The findings demonstrate that the integration of GAN inpainting with YOLOv8 effectively enhances occluded object detection in hospital environments, offering a viable solution for improved asset monitoring systems.

User-Centered Design Approach in Mobile AR: Application for Hydrocarbon Visualization in Chemistry Course

Fariz Abqari Fawwaz Illahi — 2025-04-17

This study addresses the challenges in high school chemistry education, particularly in understanding hydrocarbon compounds, by designing a Mobile Augmented Reality (MAR) application using the User-Centered Design (UCD) method. The research focuses on enhancing the visualization of submicroscopic and symbolic aspects of chemistry, which students often find abstract and complex. Through iterative design processes involving teachers and students, the study developed an interactive MAR application that displays virtual ball-and-stick models of hydrocarbon compounds. The application was evaluated using the User Experience Questionnaire (UEQ), measuring six aspects of user experience: attractiveness, perspicuity, efficiency, dependability, stimulation, and novelty. Results showed significant improvements across all dimensions from the initial to the final iteration, with five out of six scales achieving "Excellent" ratings in the final version. The study demonstrates the effectiveness of UCD in creating an engaging and user-friendly educational tool, highlighting the potential of MAR technology to address longstanding challenges in chemistry education. The positive user perceptions suggest that when designed carefully considering user needs, MAR applications can significantly enhance the chemistry learning experience for high school students.

The Evaluation of Effects of Oversampling and Word Embedding on Sentiment Analysis

Nur Heri Cahyana — 2025-04-17

Generally, opinion datasets for sentiment analysis are in an unbalanced condition. Unbalanced data tends to have a bias in favor of classification in the majority class. Data balancing by adding synthetic data to the minority class requires an oversampling strategy. This research aims to overcome this imbalance by combining oversampling and word embedding (Word2Vec or FastText). We convert the opinion dataset into a sentence vector, and then an oversampling method is applied here. We use 5 (five) datasets from comments on YouTube videos with several differences in terms, number of records, and imbalance conditions. We observed increased sentiment analysis accuracy with combining Word2Vec or FastText with 3 (three) oversampling methods: SMOTE, Borderline SMOTE, or ADASYN. Random Forest is used as machine learning in the classification model, and Confusion Matrix is used for validation. Model performance measurement uses accuracy and F-measure. After testing with five datasets, the performance of the Word2Vec method is almost equal to FastText. Meanwhile, the best oversampling method is Borderline SMOTE. Combining Word2Vec or FastText with Borderline SMOTE could be the best choice because of its accuracy score and F-measure reaching 91.0% - 91.3%. It is hoped that the sentiment analysis model using Word2Vec or FastText with Borderline SMOTE can become a high-performance alternative model.

Performance of SVM Optimized with PSO as Classification Method for Sentiment Analysis UNNESâ€™s Social Media

Miftahul Janaah — 2025-04-21

The rapid growth of Big Data, particularly from social media platforms, presents organizations with vast opportunities for extracting valuable insights. For educational institutions like UNNES, sentiment analysis can be crucial for monitoring and enhancing public perception. This research explores the application of sentiment analysis using SVM optimized by PSO to improve classification accuracy. Although SVM is widely known for its effectiveness in linearly separable data, it struggles with nonlinear data. By employing kernel functions and optimizing hyperparameters through PSO, this study aims to improve SVMâ€™s performance. The results show that the optimized SVM model with the RBF kernel and PSO achieved an accuracy of 82.05%, compared to 80.96% using standard SVM, demonstrating a 1.09% improvement. These findings indicate that PSO significantly enhances the efficiency and accuracy of SVM models in sentiment analysis, making it a powerful tool for analyzing social media data in educational contexts.

Brain Tumor Detection Through Image Enhancement Methods and Transfer Learning Techniques

Afandi Nur Aziz Thohari — 2025-04-21

A brain tumor is dangerous and must be treated immediately to prevent worsening. The identification of brain tumors can be performed by a more in-depth examination by specialists or by using artificial intelligence technology through MRI datasets. Several studies have examined how artificial intelligence could be used to find brain cancer in MRI images. The algorithm usually used is CNN with the addition of transfer learning. Previous studies have produced very high accuracy, but the accuracy value can still be improved. In this study, MRI image quality is improved as a new input for modeling. The test results show that the proposed CNN Model produces an accuracy of 98.50% on the test data. This result is higher than the baseline method of 98.45%. Analysis of other metrics, such as precision, recall, and F1-score, indicates consistent performance across classes. These findings suggest that using preprocessing to improve image quality can improve Model performance. Using CLAHE and median blur to improve image quality can improve accuracy by 14.5%. This study contributes to identifying an effective combination of Model optimization techniques for image classification tasks.

Improving the Accuracy of Concrete Mix Type Recognition with ANN and GLCM Features Based on Image Resolution

Gasim Gasim — 2025-04-23

Concrete is an essential construction material that is often used due to its strength and durability, but its mix type identification often relies on conventional methods that are less efficient and accurate. This research aims to evaluate the effect of image resolution on the accuracy of concrete mix type recognition using Artificial Neural Network (ANN) and Gray-Level Co-Occurrence Matrix (GLCM) features. The method used involves analysing concrete images at various resolutions: 200 x 200, 300 x 300, 400 x 400, 500 x 500, 600 x 600, and 700 x 700 pixels. The experimental results show that higher image resolutions tend to improve recognition accuracy. all types of image sizes using 1,250 training data and 250 test data. Image sizes of 200 x 200 and 300 x 300 pixels give low accuracy of 42% and 45% respectively, while sizes of 400 x 400 and 500 x 500 pixels show an increase in accuracy to 60.5% and 62.5%. The higher resolutions of 600 x 600 and 700 x 700 pixels produced the highest accuracy of 68% and 70%, respectively. These results indicate that larger image resolutions are able to capture more details and characteristics required for more accurate concrete mix type recognition. This research has implications for improving efficiency and consistency in concrete inspection in the construction industry through the use of AI-based image recognition methods.

Interpretation of Multi Sensor Measurement Results using Fuzzy Membership Function for Landslide Early Warning System

Erna Alimudin — 2025-05-15

Central Java has several areas prone to landslides. One of them is in Tembalang District in Semarang City, Central Java Province, Indonesia .Landslides can be caused by very high rainfall and there are no trees to support the soil, resulting in land shifting. Landslaezdide disasters are very dangerous because they cause many casualties. Therefore, there is a need for an early warning system for landslides. The landslide early warning system uses several sensors, namely rainfall sensors. Therefore, there is a need for an early warning system for landslides. The landslide early warning system uses several sensors, namely rainfall sensors, soil moisture sensors and soil movement. The sensor data will be processed using fuzzy logic so that the results can be more accurate. Early warning of landslides has several conditions, namely low risk to very high risk. Based on the results of real-time data collection in the landslide disaster early warning system, the results obtained were that the sensors were working well and communication sending data to the website was running well. Data processing has been carried out and can be processed via a controller with a fuzzy logic logic algorithm. The results obtained were that based on sensor data taken early warning of landslides still had a low risk with a value of 0.5375 and a medium risk with a value of 0.5875. This is due to moderate rainfall and high soil moisture, as well as ground movement â‰¥ 0.1

Software Effort Coefficient Optimization Using Hybrid Bat Algorithm and Whale Optimization Algorithm

Alifia Puspaningrum — 2025-05-15

Software effort estimation is a crucial aspect in software engineering, especially in project management. It defines an effort required by a person to develop an application in certain of time. One of models which widely used for this purpose is Constructive Cost Model (COCOMO) II. In COCOMO II, two coefficients have a significant role in determining the accuracy of the effort estimation. Various methods have been conducted to estimate these coefficients to closely match the actual effort with the predicted values, such as particle swarm optimization, cuckoo search algorithm, etc. However, several metaheuristics has limit in exploration and exploitation to find optimal value. To overcome this problem, a hybrid metaheuristic combining the Bat Algorithm and Whale Optimization Algorithm (BAWOA) is proposed. This approach aims to optimize the two COCOMO II coefficients for better estimation accuracy. Additionally, the proposed method is compared with several other metaheuristic algorithms using the NASA 93 datasets. There are two evaluation criteria used in this research namely Magnitude of Relative Error (MRE) and Mean Magnitude of Relative Error (MMRE). With the optimal score among comparing method. proposed method achieves superior effort estimation, with an MMRE of 51.657%. It also proves that hybrid BAWOA can estimates predicted effort close to actual effort value.

Optimization of Naive Bayes and Decision Tree Algorithms through the Application of Bagging and Adaboost Techniques for Predicting Student Study Success

Endi Febriyanto — 2025-05-15

In the Education Assessment Center of the Ministry of Education and Culture, the average grade for each student's subject in the last few years has been below 70. This condition cannot be allowed to continue. A special analysis is needed regarding factors that can help improve student grades. Predictions of student study success are urgently needed. These predictions can anticipate negative impacts that occur, including increased risk of dropout, decreased student motivation to learn, and individual potential that does not develop. The Naive Bayes and Decision Tree algorithms have been used to predict student study success. However, among its advantages, these two algorithms still have several short comings. It can cause the algorithm's performance not to be as expected. Several methods in ensemble techniques can improve algorithm performance. Two methods that are often used and can help improve the performance of classification algorithms are Bagging and Adaboost. This Study will combine Bagging and Adaboost into the Decision Tree and NaÃ¯ve Bayes algorithms to optimize the results in predicting student success. The stages carried out are initial Study, data collection, data pre-processing,data processing and evaluation model, and analysis of the results. The results show that Bagging and Adaboost techniques have been proven effective in improving accuracy, precision, recall, and F1-Score performance. Combining the naÃ¯ve Bayes algorithm with Adaboost increases accuracy, precision and recall significantly by 1.95%, 28.98%, and 15.79%.

User Satisfaction Analysis Using a Combination of End User Computing Satisfaction and Servqual on the Maluku Province KEMENAG Website

Nalbraint Wattimena — 2025-05-15

In recent years, the need for online service information has been increasing. To meet this need, the Regional Office of the Ministry of Religious Affairs in Maluku Province, through religious institutions, provides service information on their website using information system technology. In the last three months, total visits to the Ministry's website have increased by 39.4%. However, there are still issues with the website, such as complicated and time-consuming navigation due to large and numerous menu displays. A user satisfaction survey for this website has also never been conducted. Due to these issues, the researchers are interested in studying user satisfaction with the Ministry of Religious Affairs website using a combination of the EUCS and Servqual methods. This study aims to obtain an in-depth understanding of user satisfaction with the website using these combined methods and to identify various factors influencing user satisfaction. The results indicate that 8 variables have very high satisfaction values. From the T-test conducted, 7 variables affect user satisfaction: accuracy, format, ease of use, timeliness, responsiveness, assurance, and empathy. Among these, the most influential variable is format. Meanwhile, one variable, content, does not affect user satisfaction. According to the F-test results, the variables content, accuracy, format, ease of use, timeliness, responsiveness, assurance, and empathy collectively influence user satisfaction.

Optimization of Electric Multiple Unit Headway

Ruliyanta Ruliyanta — 2025-05-15

The needs of the people of JABODETABEK, Indonesia, for fast, safe, and comfortable means of transportation, such as electric rail trains, are increasing. In 2023, the narrowing of the headway on the East Bekasi-Cikarang station route will cause frequent tripping of the traction substation. This is due to the increasing frequency of multiple electric unit trips and the lack of power capacity at the traction substations to supply electrical power. In addition to that, there is a voltage drop in the overhead power network because the distance between the traction substations is too long. The fastest headway is 3 minutes from the original 13 minutes. This research aims to optimize the power capacity of the traction substation in the LAA 1.10 Cikarang area. The method is load flow analysis using ETAP 19.0.1 software. Results of the design for adding the Tambun Insertion substation and the Telaga Murni Insertion substation. On a 3-minute headway, the average voltage drop increased by 22.5% on the East Bekasi - Cibitung route from 1,222 VDC to 1,497 VDC. Meanwhile, the Cibitung-Cikarang route, originally 1,282 VDC, became 1,494 VDC, or an increase of 16.5%.

An Application Inverse Kinematic Based LQR Control for a 3 DOF Robot Arm

Adri Firmansya Sofyan — 2025-05-15

Linear Quadratic Regulator (LQR) is one of the optimal control methods on state space-based systems. The LQR control method is an option to be applied to the 3 DOF robot arm because multi-link systems such as robot arms are basically non-linear with quite complex modeling. Using conventional control methods has many trade-offs to find optimal stability between the parameters on the robot arm. System modeling is formulated using the Lagrangian dynamics and Euler-Lagrange method to obtain a nonlinear model of the system and then linearized it using Taylor series expansion. The values of the Q and R matrices can be adjusted to obtain a good system response for a particular trajectory. Tunning the Q and R parameters can also improve the stability of the system by reducing overshoot but causing the rise time of the system to increase