cover
Contact Name
Mesran
Contact Email
mesran.skom.mkom@gmail.com
Phone
-
Journal Mail Official
jurnal.bits@gmail.com
Editorial Address
-
Location
Kota medan,
Sumatera utara
INDONESIA
Building of Informatics, Technology and Science
ISSN : 26848910     EISSN : 26853310     DOI : -
Core Subject : Science,
Building of Informatics, Technology and Science (BITS) is an open access media in publishing scientific articles that contain the results of research in information technology and computers. Paper that enters this journal will be checked for plagiarism and peer-rewiew first to maintain its quality. This journal is managed by Forum Kerjasama Pendidikan Tinggi (FKPT) published 2 times a year in Juni and Desember. The existence of this journal is expected to develop research and make a real contribution in improving research resources in the field of information technology and computers.
Arjuna Subject : -
Articles 926 Documents
Komparasi Performa Klasifikasi Sentimen Masyarakat Terhadap Kurikulum Merdeka di Sekolah Menggunakan SVM dan KNN Apriyani, Risa Fitria; Megawaty, Dyah Ayu
Building of Informatics, Technology and Science (BITS) Vol 6 No 4 (2025): March 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i4.6877

Abstract

The Independent Curriculum is a strategic education policy that aims to increase learning flexibility and develop student competencies in the 21st century. This research focuses on analyzing public sentiment towards the implementation of the Independent Curriculum using two machine learning algorithms, namely Support Vector Machine (SVM) and K-Nearest Neighbors (KNN). One of the main challenges in this study is the imbalance of sentiment data that includes negative, neutral, and positive classes. To overcome this, the Synthetic Minority Oversampling Technique (SMOTE) technique was applied to balance the distribution of data between classes. The results show that the SVM method is superior to KNN with an overall accuracy of 92% and a high F1-score in the majority class (Neutral: 96%), although the performance in the minority class (Negative: 43% and Positive: 40%) still needs improvement. In contrast, the KNN method recorded a lower overall accuracy of 31% but had a more even distribution of errors. After the implementation of SMOTE, there was a significant improvement in both methods, especially in recognizing minority classes. This study concludes that SVM is more effective for sentiment classification tasks on datasets with class imbalances, and recommends further exploration of ensemble methods to improve the quality of prediction and model generalization.
Perbandingan Algoritma K-Nearest Neighbor dan Support Vector Machine Pada Pengenalan Pola Tanda Tangan Digital Yadin, Yuli; Megawaty, Dyah Ayu
Building of Informatics, Technology and Science (BITS) Vol 6 No 4 (2025): March 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i4.6982

Abstract

In the fast-paced digital era, identity security has become crucial, and digital signatures play an important role in verification and authentication. This study focuses on the analysis and comparison of the performance of the K-Nearest Neighbors (KNN) and Support Vector Machine (SVM) algorithms in digital signature pattern recognition. Both algorithms are widely used in classification tasks, and this study aims to identify which algorithm is most effective in recognizing and classifying digital signatures with the highest accuracy. Digital signature data was collected from various sources, including public datasets and directly collected data. Key features were extracted using the Gray-Level Co-occurrence Matrix (GLCM) method, which is effective in describing the texture and pattern of the signature. These features were used to train the KNN and SVM classification models. The performance of both algorithms was evaluated based on accuracy, precision, and recall metrics. The results showed that KNN with a value of k = 3 achieved an accuracy of 91.42%, while SVM with a linear kernel excelled with an accuracy of 97.06%. In addition, SVM is also more stable in handling complex signatures and has higher precision and recall than KNN, at 97.52% and 97.06%, respectively. On the other hand, KNN is faster in the training process and has a simpler implementation. This study provides valuable insights into the selection of optimal classification algorithms for digital signature recognition applications. The results of this study can be a guide for security and authentication system developers in choosing the most effective method to protect identity and prevent signature forgery.
Klasifikasi Penyakit Daun Pada Tanaman Terong dengan Metode K-Nearest Neighbors Hariansyah, Oke; Saprudin, Saprudin; Cahyono, Yono; Rosyani, Perani
Building of Informatics, Technology and Science (BITS) Vol 6 No 4 (2025): March 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i4.7016

Abstract

Eggplant (Solanum melongena) is an important agricultural commodity with high economic value. However, various leaf diseases can hinder its growth and reduce crop yields. Therefore, rapid and accurate identification and classification of leaf diseases are crucial for improving agricultural productivity. This study proposes the use of the K-Nearest Neighbors (KNN) method for classifying eggplant leaf diseases based on image analysis. The model is developed using color histogram features extracted from leaf images as the basis for classification. This research involves collecting a dataset of eggplant leaf images with various disease categories, extracting color features using RGB and HSV color models, and implementing a KNN model with k=3k=3k=3. The model's performance is evaluated using accuracy, precision, recall, and F1-score metrics. Experimental results show that the KNN model achieves an accuracy of approximately 87%, but challenges remain, such as dataset imbalance and misclassification of disease classes with similar color patterns. To improve accuracy, this study explores data augmentation techniques and optimizes the KNN model parameters. This research aims to enhance the effectiveness of KNN in detecting and classifying eggplant leaf diseases, ultimately assisting farmers in managing their crops more efficiently and effectively.
Animal Caregiver Selection by Applying ARAS Method Decision Support System and Entropy Weighting Utomo, Dito Putro; Syahrizal, Muhammad; Hondro, Rivalri Kristianto; Saputra, Imam
Building of Informatics, Technology and Science (BITS) Vol 7 No 1 (2025): June (2025)
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i1.7048

Abstract

Animal care is a profession that is responsible for taking good care of animals or checking the condition of animals for the purpose of animal health. To get a qualified animal care worker according to the company's needs, it takes quite a long time, because animal care workers who apply for a job at a company must first go through several tests in order to meet the criteria required by the company. So that animal care workers are needed and required to care for animals, maintain their health and pay attention to the nutrition of the animals. So that there is no extinction of protected animals. The current animal care worker acceptance procedure in the wildlife park is that applicants submit identity files, if they pass the applicant's files, they take an interview test and the last test, the applicant must practice in the field directly to find out how well the applicant is able to adapt to animals. In calculating the value, problems often occur in this acceptance process. A decision support system (DSS) is an interactive information system that provides information, modeling, and data manipulation. In this case, the author uses the Entropy method and the ARAS (Additive Ratio Assessment) method to solve it. The Entropy method can be used to calculate weights based on data characteristics in the criteria, the higher the variation between data in the criteria, the higher or more important the weight of the criteria. While the ARAS (Additive Ratio Assessment) method is used for ranking. The use of the Entropy method as a weighting aims to ensure that the weighting process is carried out based on objective value assignment. In the application of the Entropy Method, a weighting of the criteria value is produced where the criteria with the highest to lowest values ​​are Certificate, Work Experience, Age, Interview and Education with the highest value being 0.768 and the lowest being 0.021. Then the process of applying the selection of animal nurses using the ARAS method obtained the result that alternative A1 was selected as an animal nurse with a value obtained of 0.0816.
Nearest Neighbor Interpolation and AES Encryption for Enhanced Least Significant Bit (LSB) Steganography Anggraini, Nenny; Wardhani, Luh Kesuma; Assyahid, Muhammad Hudzaifah; Hakiem, Nashrul; Yusuf, Muhammad; Setyawan, Okky Bagus
Building of Informatics, Technology and Science (BITS) Vol 6 No 4 (2025): March 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i4.7079

Abstract

The increasing use of digital communication raises concerns about data security, especially when transmitting sensitive information. Steganography conceals messages within digital media to prevent detection. However, conventional methods face storage limitations, leading to message truncation or distortion, making hidden messages more detectable. This study proposes a combination of Nearest Neighbor Interpolation (NNI) and Least Significant Bit (LSB) steganography to dynamically expand the cover image, allowing larger encrypted messages to be embedded while maintaining image integrity. NNI was chosen over other interpolation techniques such as Bilinear and Bicubic due to its lower computational complexity and preservation of sharp edges, which minimizes blurring artifacts that could make steganographic alterations more noticeable. AES-128 encryption ensures message confidentiality before embedding. The system was developed as a web-based application to improve usability. The research followed the Waterfall Software Development Life Cycle (SDLC), and Black Box Testing validated system functionality. Testing results showed that the method successfully embedded and extracted messages without data loss, maintaining PSNR values above 40 dB, ensuring minimal perceptual distortion. However, the maximum interpolation limit was 5310 × 5310 pixels, beyond which system constraints caused failures. The stego-images retained original aspect ratios, reducing suspicion. Despite its success, the system remains vulnerable to modifications such as color changes, cropping, rotation, and compression, which can disrupt the message.
Seleksi Fitur SelectKBest Dalam Prediksi Kelulusan Mahasiswa Tepat Waktu dengan Decision Tree Putra, Yogi Erka Julyansa; Imelda, Imelda; Suryadih, Suryadih
Building of Informatics, Technology and Science (BITS) Vol 6 No 4 (2025): March 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i4.7086

Abstract

University in Jakarta are facing issues with a surge in the number of students not graduating on time within the organizational office. This recommendation analyzes the performance used to predict timely student graduation. The primary objective of this consideration is to create an estimation illustration using a Decision Tree. The data used combines information about various components. The evaluation of the emerging execution is based on metrics such as accuracy, precision, and review. Timely graduation provides various benefits to colleges. Firstly, it enhances the institution's reputation as a provider of quality instruction that supports students in completing their studies almost on time. Colleges can improve the quality and sustainability of instruction by implementing methods based on this demand. The consideration involves creating a desktop application to input substitute student data and predict whether each substitute student will graduate on time or not. This examination makes a significant contribution to efforts aimed at advancing the quality of teaching in colleges and helping substitute students better achieve their academic goals. The data, collected from universities in Jakarta, consists of 308 student records. The results almost illustrate that the model using highlight assurance yields an accuracy of 97.85%, while the model without highlight options yields an accuracy of 93.54%.
Estimasi Jarak Pandang Meteorologi di Bandar Udara Menggunakan Metode Back Propagation dan CNN Maesaroh, Siti; Muludi, Kurnia; Triloka, Joko
Building of Informatics, Technology and Science (BITS) Vol 7 No 1 (2025): June (2025)
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i1.7138

Abstract

Airports in Indonesia often face bad weather problems that affect visibility and impact flight operations. Historical data shows several incidents caused by decreased visibility due to fog or rain that resulted in flight delays and cancellations. It can be said that the importance of more accurate visibility estimates to improve safety and operational efficiency at airports. The purpose of this study was to determine the performance of the Back Propagation and Convolutional Neural Network (CNN) models in estimating meteorological visibility at airports because accurate visibility is very important in determining operational decisions, especially during bad weather conditions. The selection of the Back Propagation method is based on its advantages in handling various types of data dynamically and in a directed manner so that it is more precise in predicting visibility based on interrelated meteorological variables. While Convolutional Neural Network (CNN) is very effective in handling problems involving image data. However, currently there are quite a lot of studies that use CNN for text processing because the results are quite promising. The data used is meteorological data that includes temperature, humidity, air pressure, wind speed and other parameters at Radin Inten II Airport. From the results of this study, the Backpropagation model is better in ROC AUC (85%) compared to CNN (84%), this shows a slight advantage in distinguishing classes. The CNN model is better in Precision by 71% compared to Back Propagation 70%, which means it is slightly better at avoiding false positive predictions. CNN has a higher correlation on the test data (0.20) compared to Back Propagation (0.18) indicating its predictions are slightly more in line with the actual data. The larger correlation difference in CNN (0.18) compared to Back Propagation (0.10) indicates a higher possibility of CNN overfitting compared to BP. Since both models show almost the same performance and the difference is not too significant, the choice of model can depend on the specific needs in the implementation. If the goal is to get a more stable model, then Backpropagation is more recommended because it has a smaller correlation difference and higher ROC AUC. However, if what is sought is a model with more accurate predictions in real scenarios, then CNN can be a better choice because it has higher Precision and better test correlation.
Komparasi Algoritma Support Vector Machine dan Decision Tree Dalam Analisis Sentimen Publik Terhadap Penerapan PPN 12% Putra, Djalu Bintang; Suryono, Ryan Randy
Building of Informatics, Technology and Science (BITS) Vol 7 No 1 (2025): June (2025)
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i1.7147

Abstract

The implementation of the 12% Value-Added Tax (VAT) policy in Indonesia has generated various reactions from the public, both positive and negative. To understand public perception, researchers compared the performance of two algorithms, namely Support Vector Machine (SVM) and Decision Tree, in analyzing sentiment on social media. A total of 7,965 tweets were collected from the X (Twitter) platform using web scraping techniques and processed through several stages, including text cleaning, tokenization, stopword removal, stemming, and data balancing using the SMOTE method to improve model accuracy. The evaluation results showed that SVM achieved 80% accuracy, higher than Decision Tree, which only reached 68%. Based on these findings, it can be concluded that SVM is more effective in analyzing public sentiment regarding the 12% VAT policy. These findings can serve as a reference for the government and relevant stakeholders to better understand public opinion and design more suitable policies. This study also provides opportunities for further development by exploring other algorithms or more advanced data processing techniques to enhance the accuracy and effectiveness of sentiment analysis in the future.
Implementasi Transfer Learning pada Convolutional Neural Network dengan Arsitektur VGG dalam Klasifikasi Down Syndrome di Asia Uwar, Tarissa Rizky Salsabiila; Aditya, Christian Sri Kusuma
Building of Informatics, Technology and Science (BITS) Vol 7 No 1 (2025): June (2025)
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i1.7150

Abstract

Early detection of Down syndrome is crucial for enabling early intervention and providing healthcare education for children. Down syndrome is associated with specific facial features, such as distinct characteristics of the eyes, nose, lips, face shape, hair, and skin color, which can be analyzed using computer vision techniques. This study aims to classify Down syndrome, especially in the Asian Region, which includes countries with medium/low SDI. The study proposed a CNN based on the VGG16 and VGG19 architectures by implementing transfer learning and augmentation. Augmentation is performed to balance the number of images between classes, while transfer learning is used to train the model first on ImageNet data. The dataset used consists of two categories, Down syndrome and Healthy. The results indicate that the VGG16 model has higher sensitivity and is able to classify more cases of Down syndrome, but has a fairly large prediction error. However, VGG19 model has a better specificity value and has a smaller potential for prediction error. The best model in this study was selected based on the highest validation accuracy value, where VGG19 achieved an accuracy of 93% in its best iteration, and VGG16 achieved an accuracy of 91%. These findings suggest that the proposed models, particularly VGG19, exhibit optimal performance in classifying Down syndrome, especially in the Asian region, with a lower prediction error rate.
Integrating Support Vector Machines and Geospatial Analysis for Enhanced Tuberculosis Case Detection and Spatial Mapping Jannah, Miftahul; Jazman, Muhammad; Afdal, M; Megawati, Megawati
Building of Informatics, Technology and Science (BITS) Vol 7 No 1 (2025): June (2025)
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i1.7158

Abstract

Tuberculosis (TB) remains a significant global health problem, with Indonesia ranking third in the world in terms of TB burden. Riau Province recorded 13,007 notified TB cases in 2022 with a Case Notification Rate (CNR) of 138 per 100,000 population, still far from the national target. This study aims to develop a TB case classification system using Support Vector Machine (SVM) integrated with geospatial analysis to identify TB positive cases from screening data and visualize their spatial distribution in Riau Province. The research data was sourced from the Tuberculosis Information System (SITB) of the Riau Provincial Health Office for the period January-December 2024, covering 350 samples with demographic information, clinical symptoms, and patient risk factors. The research process includes data collection, preprocessing with Min-Max and Z-Score methods, feature extraction, modeling with SVM using various kernels (RBF, Linear, Polynomial, and Sigmoid), and geospatial visualization using Google Earth Engine (GEE). The results showed that the SVM model with Linear kernel achieved the highest accuracy of 80%, sensitivity of 100%, and specificity of 80% in detecting TB cases. Geospatial analysis successfully identified clusters of TB cases in several districts in Riau Province, with Pekanbaru City (112 cases) and Rokan Hulu (89 cases) as the main hotspots. The integration of machine learning and geospatial analysis proved effective in improving TB detection and providing a comprehensive understanding of disease spread patterns in Riau Province.