cover
Contact Name
Ricky Firmansyah
Contact Email
ricky.rym@bsi.ac.id
Phone
+6281318340588
Journal Mail Official
jurnal.informatika@bsi.ac.id
Editorial Address
Jl. Kramat Raya No.98, Kwitang, Kec. Senen, Kota Jakarta Pusat, DKI Jakarta 10450
Location
Kota adm. jakarta barat,
Dki jakarta
INDONESIA
Jurnal Informatika
ISSN : 23556579     EISSN : 25282247     DOI : https://doi.org/10.31294/ji.v4i2
Core Subject : Science,
Jurnal Informatika respects all researchers Technology and Information field as a part spirit of disseminating science resulting and community service that provides download journal articles for free, both nationally and internationally. The editorial welcomes innovative manuscripts from Technology and Information field. The scopes of this journal are: Expert System Decision Support System Data Mining Artificial Intelligence System Machine Learning Genetic Algorithms Business Intelligence and Knowledge Management Big Data the manuscripts have primary citations and have never been published online or in print. Every manuscript will be checked the plagiarism using Turnitin software. If the manuscript indicated major plagiarism, the manuscript is rejected.
Articles 10 Documents
Search results for , issue "Vol 12, No 2 (2025): October" : 10 Documents clear
SemetonBug: A Machine Learning Model for Automatic Bug Detection in Python Code Based on Syntactic Analysis Imran, Bahtiar; Riadi, Selamet; Suryadi, Emi; Zulpahmi, M.; Zaeniah, Zaeniah; Wahyudi, Erfan
Jurnal Informatika Vol 12, No 2 (2025): October
Publisher : Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/inf.v12i2.25340

Abstract

Bug detection in Python programming is a crucial aspect of software development. This study develops an automated bug detection system using feature extraction based on Abstract Syntax Tree (AST) and a Random Forest Classifier model. The dataset consists of 100 manually classified bugged files and 100 non-bugged files. The model is trained using structural code features such as the number of functions, classes, variables, conditions, and exception handling. Evaluation results indicate an accuracy of 86.67%, with balanced precision and recall across both classes. Confusion matrix analysis identifies the presence of false positives and false negatives, albeit in relatively low numbers. The accuracy curve suggests a potential overfitting issue, as training accuracy is higher than testing accuracy. This study demonstrates that the combination of AST-based feature extraction and Random Forest can be an effective approach for automated bug detection, with potential improvements through model optimization and a larger dataset.
Performance Evaluation of LSTM and GRU Models for Movie Genre Classification Based on Subtitle Dialogs Using Augmented Data and Cross-Validation Yonita Putri Utami, Ni Luh Putu; Singgih Putri, Desy Purnami; Dwi Rusjayanthi, Ni Kadek
Jurnal Informatika Vol 12, No 2 (2025): October
Publisher : Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/inf.v12i2.25897

Abstract

This study aims to evaluate and compare the performance of Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) models in classifying movie genres based on subtitle dialogs. To address data imbalance across genres, data augmentation was applied to create balanced datasets with 500 and 700 samples per genre, in addition to the original dataset. The classification models were built using Word2Vec for word embedding, followed by LSTM and GRU architectures with a single hidden layer and dropout regularization. Model performance was assessed using accuracy and further validated through 5-fold cross-validation. The best test accuracy was achieved with the dataset containing 700 samples per genre, reaching 91% for LSTM and 92% for GRU. Cross-validation showed stable performance with average accuracies of 0.68 for LSTM and 0.67 for GRU. A paired t-test analysis yielded a p-value of 0.341, indicating no statistically significant difference between the two models. These findings suggest that both LSTM and GRU are effective for genre classification based on subtitle dialogs. The use of data augmentation is a key contribution of this study, enabling improved model performance on underrepresented genres. This research supports the development of automated movie recommendation systems that utilize subtitle-based genre prediction.
Batik Pattern Classification Using Logistic Regression, SVM, and Deep Learning Features Hapsari, Ratih Addina; Yuadi, Imam
Jurnal Informatika Vol 12, No 2 (2025): October
Publisher : Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/inf.v12i2.25855

Abstract

This study presents the integration of deep learning-based feature extraction with conventional machine learning classifiers for automatically categorizing Indonesian batik patterns. The research utilizes five traditional motifs: Alas Alasan, Kokrosono, Semen Sawat Gurdha, Sido Asih, and Sido Mulyo. Feature extraction was conducted using three deep learning models: Inception V3, VGG16, and VGG19, followed by classification through Logistic Regression and Support Vector Machines (SVM), with data processing performed in Orange. Experimental results show that Inception V3 combined with Logistic Regression achieved the highest classification performance, reaching 99.2% classification accuracy and an F1-score of 0.992. These results confirm the effectiveness of deep feature embeddings in improving the automatic classification of batik motifs. The study contributes to developing intelligent classification frameworks, offering a scalable approach to cultural heritage preservation through technology. Future work will focus on enhancing feature extraction methods and expanding the dataset to address motif overlap challenges.
Evaluation of Machine Learning Algorithms for Classifying User Perceptions of a Child Health Monitoring Application Rahmawati, Eka; Wibowo, Adi; Warsito, Budi
Jurnal Informatika Vol 12, No 2 (2025): October
Publisher : Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/inf.v12i2.24639

Abstract

Child growth and development are crucial aspects that every parent should monitor carefully. Proper growth and development foster the creation of a high-quality generation for the nation’s future. Recognizing the importance of monitoring children's growth, the Indonesian Pediatric developed PrimaKu, an application designed to assist parents in tracking the children's growth and development. The application includes health guidelines, growth monitoring tools, and immunization schedules. To maximize the application’s effectiveness, it is essential to evaluate its acceptance by the community, which can be assessed through user perceptions. This study evaluates the performance of machine learning algorithms, including Random Forest, Support Vector Machine, Naive Bayes, and Decision Tree, in classifying user perceptions of the PrimaKu application. The results revealed that the Support Vector Machine model achieved the highest accuracy of 81%, followed by Random Forest at 77%, Decision Tree at 74%, and Naive Bayes at 73%. Precision, recall, and F1-score used to validate the models' performance as the evaluation metrics. The findings underscore the potential of machine learning techniques in effectively classifying user feedback, providing valuable insights for improving application development and enhancing user satisfaction. This study contributes to understanding user acceptance of digital tools for child health monitoring, paving the way for better application usability and community impact
Comparative study of DistilBERT and ELECTRA-Small Models in Spam Email Classification Agusman, Ferdy
Jurnal Informatika Vol 12, No 2 (2025): October
Publisher : Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/inf.v12i2.25528

Abstract

Spam email detection is one of the challenging tasks in cybersecurity due to the variability of spam content. These characteristics make it harder to identify spam, therefore researchers create different spam detection methods. Among these, Natural Language Processing (NLP) and machine learning techniques have shown outstanding results in classifying emails as spam or non-spam. Transformer-based models, such as BERT, have demonstrated pinpoint accuracy in text classification tasks. However, the computational requirements and resources are not practical in resource-limited environments. In order to mitigate this, smaller and more lightweight models, such as the DistilBERT and ELECTRA-Small, have been developed. Both models are renowned for their efficiency and accuracy. This study focuses on the comparison of these models in terms of accuracy, precision, recall, and F1 score. Experimental results revealed that while both models excel in binary classification, notable differences emerge. ELECTRA-small shows exceptional accuracy, precision and faster processing time, while DistilBERT demonstrates superior recall, highlighting its effectiveness in minimizing false negatives.
Opinion Mining on Spotify Music App Reviews Using Bidirectional LSTM and BERT Arsi, Primandani; Firmanda, Reza Arief; Prayoga, Iphang; Subarkah, Pungkas
Jurnal Informatika Vol 12, No 2 (2025): October
Publisher : Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/inf.v12i2.25323

Abstract

The increasing number of user reviews on digital music platforms such as Spotify highlights the importance of sentiment analysis to better understand user perceptions. This study aims to develop a sentiment classification model for Spotify user reviews using a Bidirectional Long Short-Term Memory (BiLSTM) approach combined with BERT embeddings. The dataset consists of multilingual user reviews collected from the Google Play Store. Preprocessing steps include text cleaning, tokenization, and padding. BERT is utilized to generate contextual word embeddings, which are then processed by the BiLSTM model to classify sentiments as either positive or negative. The model’s performance is evaluated using a confusion matrix with accuracy, precision, recall, and F1-score metrics. The results show that the BiLSTM-BERT model achieves an F1-score of 0.8852, a recall of 0.9396, a precision of 0.8375, and an accuracy of 0.8374. These findings demonstrate the model’s effectiveness in handling multilingual sentiment analysis tasks, offering valuable insights for developers in enhancing user experience through data-driven decision-making.
Academic Information System on The Prince Jayakarta Junior High School Student Assessment System Haq, Fesa Asy Syifa Nurul; Muliati, Vika Febri; Rahman, Abdu; Zarlis, Muhammad; Darusalam, Ucuk; Siregar, Nurhayati
Jurnal Informatika Vol 12, No 2 (2025): October
Publisher : Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/inf.v12i2.26261

Abstract

Information technology has supported the development of quality school services throughout the world. However, there are still many schools that have not used it optimally, especially in Indonesia, for example at Prince Jayakarta Bekasi Middle School. As in general, schools only use Ms. Word and Ms. applications . Excel. This resulted in different assessment formats and errors when filling in grades into the report card format. The academic information system application developed in this study uses the PHP, HTML, and MySQL programming languages and is named SIAP, which means eighteen academic information systems. The purpose of making this application is so that students / parents of students can receive school assessment information in a precise, fast and accurate manner. Teachers can also use the facility to process student scores so that they are well integrated and summarized as data for the Principal to make policies. This application can be opened on any browser platform, making it easier for users to access it anywhere and anytime.
Pose Analysis and Classification in Shooting Sport Using Convolutional Neural Network and Long Short-Term Memory Sobari, Bahar; Moedjiono, Moedjiono; Rizkiawan, M. Asep
Jurnal Informatika Vol 12, No 2 (2025): October
Publisher : Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/inf.v12i2.25566

Abstract

Shooting sport requires high accuracy and speed, making training evaluation essential for athlete performance improvement. Conventional evaluation methods are often limited, thus the application of Artificial Intelligence (AI) and Computer Vision provides an effective alternative. This research aims to analyze and classify shooting sport poses using Deep Learning methods. A dataset consisting of several thousand pose images was collected from both field recordings and publicly available sources, followed by preprocessing for coordinate extraction. Convolutional Neural Network (CNN) was employed to extract coordinate data from shooting pose images, while Long Short-Term Memory (LSTM) was applied for pose classification. Experimental results demonstrated 94% accuracy, 95% Percentage of Correct Keypoints (PCK), and 4 mm Mean Per Joint Position Error (MPJPE), with training conducted at a learning rate of 0.0001 over 150 epochs on 5% test data, involving a total of 596,642 parameters. These results indicate that the proposed CNN–LSTM model provides a reliable approach for pose analysis and classification in shooting sport. The contribution of this study lies in presenting a novel dataset and framework for AI-based shooting sport evaluation, which can enhance training feedback and broaden AI applications in sports. 
Performance Evaluation of RESTful API in Sales Target Monitoring System for Direct Sales and Sales Canvassers Widiono, Suyud; Friwaldi, Restian Dwi; Anggara, Afwan
Jurnal Informatika Vol 12, No 2 (2025): October
Publisher : Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/inf.v12i2.25747

Abstract

In an increasingly competitive digital era, manual sales target monitoring often leads to delayed information and inefficiency in decision-making. This research aims to develop a web and mobile-based sales target monitoring system integrated with RESTful API to enhance the efficiency of monitoring the performance of direct sales and sales canvassers. The system is developed using the Laravel framework for the back-end and Flutter for the mobile application, with Agile methodology applied in the development process. Testing is conducted using the Black Box Testing method to ensure the accuracy of system functionalities, including user authentication, sales data management, and sales target monitoring. Additionally, load testing is performed using Apache JMeter with scenarios of 500, 750, and 1000 users. The test results show that the system has stable performance with an average response time of 758 ms for 500 users, 762 ms for 750 users, and 880 ms for 1000 users, all below the threshold of 900 ms. The error rate is recorded at 0.00%, and the system throughput exceeds the set target, indicating the system's reliability in handling simultaneous user requests. The conclusion of this research shows that the implementation of RESTful API in the sales monitoring system can improve operational efficiency, enable real-time data exchange, and support faster, data-driven decision-making. As a recommendation, further development could include broader integration with mobile applications and the implementation of AI-based analytics for sales strategy optimization.
Comparative Optimization of EfficientNetB3, MobileNetV2, and ResNet50 for Waste Classification Agustiani, Sarifah; Haryani, Haryani; Junaidi, Agus; Putri, Rizky Rachma; Emiliana, Meutia Raissa
Jurnal Informatika Vol 12, No 2 (2025): October
Publisher : Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/inf.v12i2.27533

Abstract

Waste management has become a critical challenge in efforts to maintain environmental sustainability and public health. Poorly managed waste can cause environmental pollution, reduce quality of life, and complicate recycling processes. To address this issue, this study aims to classify waste based on images while optimizing several deep learning architectures, namely EfficientNetB3, MobileNetV2, and ResNet50, to identify the best model for waste classification. The research methodology includes data collection, preprocessing, data augmentation, model development, and performance evaluation using accuracy, precision, recall, and F1-score metrics. The dataset, obtained from the Kaggle platform, consists of 4,650 images divided into six categories: battery, glass, metal, organic, paper, and plastic. The results show that EfficientNetB3 with the Adam optimizer achieved the best performance, with accuracy, precision, recall, and F1-score all at 93%, followed by ResNet50 at approximately 91%, and MobileNetV2 ranging from 70–73% depending on the optimizer. The use of different optimizers was found to influence model performance, and data augmentation helped improve generalization, especially for classes with limited samples. Limitations of this study include the relatively limited dataset coverage. Future research is recommended to expand the dataset and explore alternative or hybrid architectures. These findings demonstrate the potential of deep learning–based systems in supporting sustainable waste management.

Page 1 of 1 | Total Record : 10