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
Klasifikasi Penyakit Pada Daun Kopi Robusta Menggunakan Arsitektur AlexNet dan Xception dengan Metode Convolutional Neural Network Ashari, Nadia; Avianto, Donny
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6109

Abstract

Diseases on the leaves of robusta coffee plants can have a significant impact on the growth and yield of robusta coffee plants. The leaves of the robusta coffee plant are susceptible to various types of diseases caused by fungi, bacteria or insects with symptoms such as brown, yellow or black patches and discoloration on the surface of the leaves of the robusta coffee plant. Early detection of diseases in robusta coffee leaf plants is very important to obtain effective control to maintain plant health. In this study, a disease classification model on the leaves of robusta coffee plants was made using the Convolutional Neural Network (CNN) architecture. The architecture used in this study is AlexNet and Xception. In this study, a dataset of images of robusta coffee leaves obtained through direct observation of robusta coffee plantations in Temanggung Regency was used. The number of datasets used was 1400 data which was divided into 4 classes, namely healthy, root down, leaf rust and red spider mites. The CNN model was tested by setting parameters consisting of batch size, drop out, learning rate, optimizer and the number of epochs that varied 35, 50 and 100. The results of this study show that the AlexNet architecture model with 50 epoch tests obtains the best accuracy of 98.57% and the Xception architecture obtains an accuracy of 100% in each epoch test. Overall, the use of AlexNet and Xception architectures is very effective in classifying diseases in robusta coffee leaves, but the Xception architecture is superior in the ability to classify complex datasets and higher accuracy.
Penerapan Metode Supervised Learning dan Teknik Resampling untuk Prediksi Penipuan Transaksi Keuangan Constancio, Elven; Tania, Ken Ditha
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6110

Abstract

Financial transaction fraud can result in devastating consequences for the stability of companies, as well as huge losses for shareholders, the industry, and even the market as a whole. As fraud in financial transactions increases, there is a need for effective methods to accurately detect and prevent fraudulent activities. This study aims to compare the performance of five machine learning models, namely Random Forest, K-Nearest Neighbors (KNN), Decision Tree, XGBoost, and Extra Trees, in detecting financial transaction fraud using an imbalanced dataset. To overcome the data imbalance problem, three resampling techniques are applied, namely Synthetic Minority Oversampling Technique (SMOTE), Adaptive Synthetic Sampling (ADASYN), and Undersampling. Experiments were conducted with two training and test data sharing ratios, namely 70:30 and 80:20. The evaluation results showed that the XGBoost model was the most consistent, with the highest ROC AUC value of 99%, especially after the application of resampling techniques. The 80:20 data ratio resulted in a more balanced distribution and better model performance in detecting the minority class, particularly after resampling. This study concludes that the XGBoost model with resampling techniques is highly effective in addressing data imbalance.
Analisis Sentimen Terhadap Ulasan Pengguna Pada Aplikasi Traveloka Menggunakan Metode Naïve Al Hakim, Muchammad Gamma; Irwiensyah, Faldy
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6119

Abstract

The proliferation of user-generated reviews on digital platforms provides in-depth information to improve services. The purpose of this study is to apply the Naïve Bayes approach to analyze the sentiment of user evaluations of the Traveloka application sourced from the Google Play Store. Through online search, 10,000 evaluations were collected. Case folding, stopword elimination, tokenizing, and stemming are some of the pre-processing techniques used. Based on the review scores, the sentiment data was classified into two groups: positive and negative. Furthermore, the Naïve Bayes model was used for classification, and a confusion matrix was used to assess the results. The results showed an accuracy of 89.35%, precision of 88.44%, recall of 95.05%, and F1-Score of 91.62%. These results demonstrate the effectiveness of the Naïve Bayes approach in categorizing user reviews, providing Traveloka with important information about customer perceptions and how to improve their service quality. The findings from this study are expected to be the basis for future advancements in sentiment analysis on travel and accommodation-related applications.
Implementation of Item-Based Collaborative Filtering Algorithm for Blangkon Product Recommendation on Web-Based E-commerce System Atmojo, Cahyo Tri; Susanto, Ajib
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6120

Abstract

In the development of technology at this time, especially in the trade sector, there is no escape from the development of information technology which has had a significant impact. The most obvious form in the development of information technology in the trade sector is e-commerce, which allows transactions between sellers and buyers to be easier. Not only that, the problem now is that users must be spoiled with features that help to recommend user desires. This requires a recommendation system to help select user desires based on products with high ratings. Therefore, it must continue to develop a system that has features to support the sales system. To achieve the system needs to require a method that supports such as using the collaborative filtering method. This method focuses the analysis on similarities between items, because it is more stable and not always sensitive to changing data with a large number of users. The collaborative filtering method is used in the recommendation system to predict inter-user preferences for blangkon products based on the similarity of other user patterns, so that product recommendations appear that they have never seen or bought before. This technique uses an item-based model in it. The results of the performance test to determine the level of prediction accuracy of the method in this study using the mean absolute error. With MAE for three times trying to get a value of 0.5, 0.3 and 0.2.
Analisa Optimasi Grid Search pada Algoritma Random Forest dan Decision Tree untuk Klasifikasi Stunting Rahmayani, Ririt Sheila Tina; Budiman, Fikri
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6128

Abstract

Stunting is a serious problem that is of global concern because of its significant impact on the health and growth of children under five. This condition occurs due to long-term malnutrition. In Indonesia, nutritional problems are still common, including stunting which affects children's growth and development. In this regard, data mining has an important role in facing this challenge. Therefore, the aim of this research is to optimize stunting classification using Decision Tree and Random Forest algorithms optimized with Grid Search. This optimization was carried out to increase the accuracy of the two algorithms and identify algorithms that are superior in determining stunting. The dataset used consists of 10,000 toddler data with important attributes related to health conditions. The analysis results show that the initial Decision Tree model has an accuracy of 70.2%. After optimization using Grid Search, the accuracy of the Decision Tree model increased significantly to 82.8%. Meanwhile, the initial Random Forest model achieved an accuracy of 77.9%, and after optimization with Grid Search, its accuracy increased even higher compared to Decision Tree, namely 84.1%. This increase reflects the effectiveness of optimization in increasing the model's ability to classify stunting more accurately. This research provides important insights into the effectiveness of both algorithms in identifying stunting and emphasizes the importance of optimization to improve classification accuracy, which can support appropriate interventions for the well-being of future generations.
Penerapan Algoritma Naïve Bayes pada Analisis Sentimen Aplikasi Traveloka pada Platform Playstore Putri, Eka Ardiya; Berlilana, Berlilana
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6130

Abstract

The number of internet users in Indonesia is increasing every year, making it the fastest-growing country in the world, next only to China, India and the United States. In 2017, in Indonesia, the digital economy sector had a high impact on GDP, showing a figure of 7.3%, while the total economic development only reached 5.1%. Traveloka appeared in 2012 and has grown rapidly to be classified as the most superior travel application in Southeast Asia. As applied by the Traveloka application, it applies data scraping to collect 5000 review data from the intended platform. With the increase of Traveloka app user reviews on Playstore, the main challenge is to classify the sentiment of the reviews automatically and accurately. The purpose of this research is to find out the extent of user assessment of the Traveloka application. The results show that the model has an Accuracy of 0.91, indicating that 91% of the total data was predicted correctly. The model'sF1 Score of 0.90 reflects the optimal balance between Precision and Recall, indicating that the model is not only correct in predicting positive results, but also able to capture almost all positive examples. Precision of 0.92 indicates a high level of accuracy in positive predictions, while Recall of 0.88 indicates that the model's ability to detect all positive data is very good. In this analysis, out of the 940 data used, 250 True Positive (TP), 18 False Positive (FP), 608 True Negative (TN) and 64 False Negative (FN) were found, with an 80:20 data split. The findings show that the model can predict most of the data accurately, despite some errors in positive and negative classification. These results indicate that the model has high effectiveness in the identification and prediction of positive data, providing a strong basis for further applications in data analysis.
Optimasi Data Preprocessing dan Hyperparameter Tuning pada Klasifikasi Penyakit Daun Apel menggunakan DenseNet169 Ramadhan, Gilang Satria Putra; Maimunah, Maimunah; Nugroho, Setiya
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6134

Abstract

Apples are one of the horticultural commodities in Indonesia with production reaching 5,235,955 quintals in 2022, but decreasing to 3,925,628 quintals in 2023. One of the causes of this decline is diseases in apple plants that occur on the leaves, such as scab, black rot, and cedar rust, which can result in a decrease in the quality and quantity of production. Therefore, technology is needed for fast and accurate classification of diseases on apple leaves. This study uses a CNN model with DenseNet169 with optimization on data preprocessing and hyperparameter tuning to improve the accuracy of the apple leaf disease classification model. A total of 36 combinations of data preprocessing and hyperparameter tuning scenarios were tested on the apple leaf image dataset consisting of 4 classes: scab, black rot, cedar rust, and healthy. The optimal scenario is obtained from a combination of RGB + CLAHE with RMSprop optimizer and a learning rate of 0.0001 (P6 + H4), which results in 99.39% accuracy, 99.4% precision, 99.39% recall, and 99.39% f1-score. The results of this study show that the selection of the right preprocessing data and hyperparameter tuning greatly affects the performance of the apple leaf disease classification model.
Implementasi Algoritma YOLO V8 (You Only Look Once) Dalam Deteksi Penyakit Daun Durian Putra, Raihan Restu; Maimunah, Maimunah; Sasongko, Dimas
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6136

Abstract

The durian plant (Durio zibethinus Murr.) is one of the leading tropical fruit agricultural products in Indonesia with high economic value. However, durian productivity is often disrupted by leaf disease attacks such as Algal Leaf Spot, Leaf Blight, and Leaf Spot, which results in a decrease in the quality of the crop and its quantity. As a step to address this problem, the goal of this study is to automatically detect durian leaf disease using the YOLOv8 algorithm, a new deep learning model developed to detect objects directly in real time. This study used a data set that included 420 images of durian leaf disease in four categories, Algal Leaf Spot, Leaf Blight, Leaf Spot, and No Disease. This study uses three dataset distribution scenarios (70:20:10, 75:15:10, and 80:10:10), with various epoch and batch size configurations. The results show that the 70:20:10 scenario with 100 epochs and batch size 16 produces the best performance, with a precision value of 0.994, recall of 0.989, mAP50 of 0.990, and mAP50-95 of 0.927. The model developed is able to detect durian leaf disease with high accuracy and fast inference time. The implementation of this model through the Roboflow platform allows efficient use and is ready to be implemented to support the sustainable increase in durian productivity. This research also makes a significant contribution to the development of deep learning-based agricultural technology in Indonesia.
Klasifikasi Kanker Kulit menggunakan Convolutional Neural Network dengan Optimasi Arsitektur Sinaga, Jesica Trivena; Faudyta, Haniifa Aliila; Subhiyakto, Egia Rosi
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6141

Abstract

Skin cancer is a severe condition characterized by the abnormal growth of skin cells, often triggered by ultraviolet exposure and genetic factors. Early detection of skin cancer is essential for improving patient recovery rates, given the high incidence and significant impact of the disease. This study aims to develop a skin cancer classification system using the Convolutional Neural Network (CNN) method with the VGG-16 architecture, known for its effectiveness in medical image analysis. The CNN method was chosen because it can extract complex features from images. At the same time, the VGG-16 architecture was selected for its depth and ability to capture fine details in images—critical for distinguishing between types of skin cancer. The dataset was sourced from the ISIC platform and optimized through data augmentation techniques to address data imbalance issues. The research results indicate that while a basic CNN can provide good accuracy, implementing the VGG-16 architecture significantly increases accuracy. The basic CNN model achieved a training accuracy of 95.68% and a validation accuracy of 89.83%, whereas the CNN with VGG-16 reached a training accuracy of 96.21% and a validation accuracy of 90.89%. These findings suggest that combining CNN with VGG-16 effectively detects skin cancer, with VGG-16 providing a slight accuracy improvement, highlighting this architecture's potential as a more accurate tool to support skin cancer diagnosis.
Analisis Sentimen Komentar Media Sosial Twitter Terhadap Tes CPNS dengan Algoritma Naive Bayes Nurhidayat, Rifki; Hendrastuty, Nirwana
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6148

Abstract

The Calon Pegawai Negeri Sipil (CPNS) is one of the most sought-after careers in Indonesia, with the number of applicants increasing every year. The CPNS selection process attracts public attention and triggers various opinions, both positive and negative, which are widely conveyed through social media such as Twitter. This research aims to analyze public sentiment towards the CPNS selection process using the Naive Bayes algorithm. The data used in this study consists of 5,599 comments on Twitter, with a composition of 5,269 negative sentiment data and 323 positive sentiment data. Tests were conducted using several data sharing ratios, namely 80:20, 70:30, 90:10, and 50:50. The results show that the 70:30 ratio provides the best accuracy, which is 95%. However, data imbalance causes the model to focus more on negative sentiment. To address this issue, the Synthetic Minority Over-sampling Technique (SMOTE) was applied, which successfully improved the model's performance in classifying positive data, with precision and recall reaching 85-98%. After the application of SMOTE, the overall accuracy decreased slightly to 91% at 80:20, 70:30, and 90:10 ratios, but the model became more effective in detecting both sentiments. The results of this study provide insight into the public's views on CPNS selection and can be used by the government to improve the selection process in the future. With this approach, it is expected that government agencies can better understand public perceptions and optimize a more transparent and fair recruitment system.