cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
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
Published by Forum Kerjasama Pendidikan TInggi
ISSN : 26848910     EISSN : 26853310     DOI : -
Core Subject : Science,
Computer Science & IT
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 889 Documents
 81   82   83   84   85 
Peramalan Multivariat Saham Bank Indonesia dengan Model ARIMA dan LSTM Ramadhan, Akhdan Ferdiansyah; Agastya, I Made Artha
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.v7i1.7352

Abstract

Stock price forecasting is a crucial aspect of financial market analysis, particularly in supporting more accurate and informed investment decision-making. This study compares the performance of the statistical Autoregressive Integrated Moving Average (ARIMA) model with three variants of the Long Short-Term Memory (LSTM) architecture, namely Vanilla LSTM, Bidirectional LSTM, and Stacked LSTM, in predicting closing prices and trading volumes of Indonesian bank stocks—specifically BBCA.JK, BBRI.JK, and BMRI.JK. The data were obtained from Kaggle and processed through normalization, transformation, and model training stages using Google Colab and TensorFlow. Evaluation was conducted using RMSE, MAE, and MAPE metrics. The results indicate that ARIMA performs better in forecasting closing prices, achieving an average MAPE of 1.9%, while Bidirectional LSTM yielded the best results in forecasting trading volumes, particularly for BBRI and BMRI stocks. However, the prediction error for volume data remains relatively high (average MAPE of 36.4%) due to its volatile nature. These findings suggest that data characteristics significantly influence model effectiveness. LSTM-based models demonstrate superior capabilities in capturing complex non-linear patterns and exhibit advantages in multivariate forecasting compared to the ARIMA model. This study is expected to serve as a reference for selecting appropriate forecasting models in the context of Indonesian banking stock markets. The results highlight a trade-off between ARIMA, which excels in modeling linear patterns such as closing prices, and LSTM, which is more adaptive to non-linear patterns like trading volumes.
Performance Evaluation of Deep Learning Models for Cryptocurrency Price Prediction using LSTM, GRU, and Bi-LSTM Yanimaharta, Arya; Santoso, Heru Agus
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.v7i1.7353

Abstract

Cryptocurrency price prediction poses a significant challenge in the digital finance landscape due to its high volatility and complex data patterns. Traditional statistical methods often fail to capture the nonlinear and temporal dependencies inherent in cryptocurrency price movements. This study addresses this issue by evaluating and comparing the performance of three deep learning architectures, namely Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and Bidirectional LSTM (Bi-LSTM). in predicting the closing prices of Bitcoin (BTC), Ripple (XRP), and Dogecoin (DOGE). The dataset was obtained from Yahoo Finance, covering the period from January 1, 2020, to April 30, 2025. The models were evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Symmetric Mean Absolute Percentage Error (SMAPE), with a forecasting horizon of 30 days. The results of this study indicate that the LSTM model achieved the highest accuracy for Bitcoin and Ripple, with MAPE values of 2.58% and 4.33%, respectively. Meanwhile, the GRU model demonstrated the best overall performance for Dogecoin, with RMSE (0.0131), MAE (0.0084), MAPE (4.12%), and SMAPE (4.06%). On the other hand, the Bi-LSTM model exhibited the lowest performance across all tested cryptocurrencies. These findings highlight the importance of selecting an appropriate model for developing accurate cryptocurrency price prediction systems. This study contributes to the field by providing a detailed comparative analysis of model performance across cryptocurrencies with differing volatility patterns, offering insights for developing more robust and tailored predictive systems in volatile financial environments.
Analisis Seleksi Penerimaan Karyawan Bank Menggunakan Metode Simple Additive Weighting (SAW) dan Pembobotan Rank Order Centroid (ROC) Naibaho, Sri Fitriani; Bangun, Budianto; Masrizal, Masrizal
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.v7i1.7360

Abstract

This study aims to improve the efficiency and effectiveness of the employee recruitment selection process at Bank BRI Aek Kanopan. The main problem faced is the lack of transparency and systematicity in decision making, so an objective approach is needed to determine the best candidate. As a solution, this study uses the Simple Additive Weighting (SAW) method, which allows structured weighting of criteria, such as last education, technical skills, work experience, communication skills, and age. The SAW method is applied to normalize data and compare candidates based on preference values ​​(Vi). This study involved 25 alternative candidates as evaluation objects, with the normalization and weighting process carried out based on predetermined criteria. The normalized data are presented in a table, with preference values ​​ranging from 0.667 to 0.793, which indicates the level of suitability of candidates to the selection criteria. The results of the study indicate that the SAW method is effective in determining the best candidates, increasing transparency, and facilitating decision making by management. Based on these findings, Bank BRI Aek Kanopan is advised to adopt the SAW method routinely, train staff involved in the selection process, and conduct periodic evaluations to maintain the effectiveness of this method. This research provides an important contribution in the development of human resource management, especially in improving the efficiency and accuracy of the employee recruitment selection process.
Sentiment Analysis Aplikasi Mobile TIX ID di Playstore Menggunakan Algoritma Random Forest Ramadhini, Reffina; Sanjaya, M Rudi; Ruskan, Endang Lestari; Indah, Dwi Rosa
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.v7i1.7361

Abstract

TIX ID is one of the e-ticketing entertainment platforms for film orders that has experienced a rapid surge in Indonesia. The various features offered by the TIX ID application must of course be able to meet user expectations in order to compete in the market. The influence of reviews provided by users has a very important impact on the reputation of an application, whether it is positive reviews in the form of text and then processed into negative review information. Sentiment analysis is a study used in analyzing a review or perspective whose final result is in the form of positive or negative text information. The research that has been carried out using the Random Forest algorithm has succeeded in collecting review data of 2000 samples labeled positive and negative. Random Forest modeling in the study used the evaluation of the confusion matrix model and classification report which managed to achieve an accuracy of 87%, performance in the negative class showed high precision of 85%, negative recall rate of 92%, and f1-score of 88%. Then in the positf precision class reached 91%, recall was 83%, and f1-score was 87%. While the macro average and weighted average values for all metrics were 88%, indicating a balance of classification performance among the classes. Overall, the application of the Random Forest algorithm model provides accurate results and makes sentiment analysis a tool that helps developers understand user satisfaction and needs on the TIX ID application.
Analisis Performa Model BiLSTM dan CNN-LSTM Dalam Prediksi Sea Water Level Pada Pelabuhan Berdasarkan Data Historis Kristyanto, Andi; Chairani, Chairani; Sriyanto, Sriyanto
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.v7i1.7364

Abstract

Indonesia is a country dominated by waters, so data on sea level rise, one of maritime weather is important. The Meteorology, Climatology, and Geophysics Agency one of its duties, namely conducting observations in meteorology. The Merak-Bakauheni Port serves the busiest crossing route in Indonesia and connects the islands of Java and Sumatra. If there is a disruption due to meteorological factors, shipping and sea transportation activities will be hampered and disrupted. The purpose of this study is to compare the performance of the BiLSTM and CNN-LSTM models in estimating sea water levels at Merak Port based on the results of the parameter analysis used. The steps begin with collecting, processing data, training the model, and analyzing the model. The data used is daily sea water level data over a period of six years from 2019 to 2024. Evaluation of MSE, MAE and RMSE values ​​is used to see the performance of the two models. From this study, the BiLSTM model produced values ​​of 0.0026 (MSE), 0.0224 (MAE), and 0.0512 (RMSE), the CNN-LSTM model values ​​of 0.0044 (MSE), 0.0319 (MAE), and 0.0664 (RMSE), it can be seen that BiLSTM method has more optimal in predicting sea water levels of Merak Port.
Analisis Kinerja Model Support Vector Machine dalam Prediksi Kasus HIV di Indonesia Berdasarkan Data Time Series Erza, Muhammad Al-Ghifari; Prasetyaningrum, Putri Taqwa
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.v7i1.7365

Abstract

Accurate predictions of HIV cases are crucial in efforts to control the epidemic effectively in Indonesia. As the number of cases and the complexity of transmission factors increase, machine learning-based prediction methods are becoming increasingly relevant. This study analyzes the performance of the Support Vector Machine (SVM) model in forecasting the number of HIV cases in Indonesia using time series data from 2012 to 2024. The CRISP-DM methodology is used as the framework for the analysis process, starting from business understanding to model deployment. The dataset used includes secondary data from the Ministry of Health, such as SIHA, national surveillance, and reports from the Directorate General of Disease Prevention and Control (Ditjen P2P). The SVM model is selected due to its ability to handle non-linear data and limited data sizes, as well as its resilience to overfitting. Model evaluation is performed using MAE, RMSE, and MAPE metrics. The results of the study show that the SVR model with an RBF kernel provides good prediction accuracy, with MAE values of 691.34, RMSE of 823.11, and MAPE of 13% on the test data. Therefore, SVM can be an effective tool to support data-driven decision-making in HIV control efforts in Indonesia.
Analisis Sentimen Publik Terhadap Program KIP-Kuliah Menggunakan Algoritma Random Forest pada Media Sosial X Rosdiyanah, Rosdiyanah; Lestarini, Dinda
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.v7i1.7366

Abstract

KIP-Kuliah program or The Indonesia Smart College Card (KIP-K) is a funding assistance from the government for students with economic difficulties who want to experience educational opportunities in higher education. This program can’t be separated from public discussion, especially regarding the issue of misuse of funds by recipients, inconsistency in fund disbursement and falsification of registration files. These problems make the public view that the KIP-K program is often still misdirected. The research aims to examine public sentiment or perception towards the KIP-K program using Random Forest algorithm combined with Word2Vec as a word weighting technique and Random Oversampling (ROS) as a balancing technique to overcome data imbalance. The dataset obtained comes via platform X or Twitter) a total of 4423 tweets with the keywords “kip-k” or “kipk” and with a vulnerable time during 2024. The model’s performance demonstrated a high accuracy of 96,57%, precision, recall, and f1-score at the same value of 97%. The results indicate that the model is effective in analyzing sentiment accurately and maintaining a balanced performance between the two sentiment classes. Based on research in this study, the Random Forest algorithm combined with Word2Vec and Random Oversampling (ROS) can produce high accuracy and can overcome data imbalance.
Segmentasi Produk Pakaian Menggunakan Algoritma K-Means Clustering dan Particle Swarm Optimization untuk Strategi Pemasaran Putra, Rio Aji Hadyanta; Prasetyaningrum, Putri Taqwa
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.v7i1.7367

Abstract

This research aims to analyze product segmentation in the apparel industry using the K-Means Clustering algorithm optimized with Particle Swarm Optimization (PSO) to generate accurate product segmentation that can support more effective marketing strategies for a company. The data used in this analysis were obtained from sales transactions of a clothing manufacturing company that offers various categories of apparel products. The dataset consists of 333 rows and includes transaction numbers, product types, quantities sold, and total sales values. The data were processed using the Python programming language via Visual Studio Code. The segmentation process was initially performed using the K-Means algorithm to group products, and the Elbow method was applied to determine the optimal number of clusters. The number of clusters obtained from the Elbow method was then optimized using PSO to find more optimal cluster counts and centroids. Cluster evaluation was conducted by comparing the values of several metrics, including the Davies-Bouldin Index (DBI), Silhouette Score, Sum of Squared Error (SSE), and the SSW/SSB ratio. Although the DBI increased slightly from 0.6690 to 0.6878, indicating greater similarity between clusters, the improvement in the Silhouette Score from 0.5513 to 0.5569 suggests better internal consistency within the clusters. Furthermore, the reduction in SSE from 418.52 to 313.25 indicates a tighter distribution of data within clusters, while the significant decrease in the SSW/SSB ratio from 0.4582 to 0.3075 demonstrates more clearly defined cluster boundaries and improved separation. The results of the study produced four distinct product clusters, enabling the company to implement more targeted and differentiated marketing strategies.
Evaluasi Performa Rmixmod dan KAMILA dalam Pengelompokan Perguruan Tinggi di Indonesia Berdasarkan Data Capaian Kinerja Bertipe Campuran Santoso, Andrianto; Kurnia, Anang; Hamim Wigena, Aji
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.v7i1.7376

Abstract

Clustering is a technique for grouping objects based on their similarities within clusters and their differences across clusters. In real-world, objects often have characteristics represented by a combination of numerical and categorical variables, requiring clustering techniques that can process mixed-type data. Model-based clustering is one of the approaches that can be utilized for such data. This study evaluates and compares two model-based clustering algorithms for mixed data type, Rmixmod, which employs a mixture model with maximum likelihood estimation and expectation-maximization, and KAMILA, which utilizes a semi-parametric approach. Both algorithms are implemented to cluster Indonesian higher education institutions based on their performance. The optimal number of clusters is determined using the Bayesian Information Criterion and the Silhouette Coefficient. Algorithms performance is evaluated using the Silhouette Coeeficient, the Calinski-Harabasz Index, and the Davies-Bouldin Index. The research results showed that the Rmixmod algorithm outperformed KAMILA in clustering Indonesian higher education institutions, with a Silhouette Coeeficient of 0.2878, a Calinski-Harabasz Index of 253.9433, and a Davies-Bouldin Index of 1.5321. The optimal number of clusters formed was five. Cluster interpretation is conducted by analyzing the mean values of PC and the distribution of categorical variables within each cluster. The clustering results are expected to serve as a foundation for the government in formulating strategic policies that are both effective and differentiated according to the characteristics of each group of higher education institutions.
Klasifikasi Kanker Payudara Berdasarkan Gambar Histopatologi Menggunakan Metode Convolutional Neural Network Dengan Arsitektur VGG-16 Nandasari, Dayang; Pratama, Irfan
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.v7i1.7377

Abstract

Breast cancer is one of the deadliest diseases with a high prevalence worldwide, especially in women. Breast cancer is the third leading cause of death in Indonesia. Based on Globocan Center data, there will be approximately 408,661 new cases and nearly 242,099 deaths in Indonesia by 2022. Early detection through histopathology images is very important to increase the patient's chances of recovery. However, the diagnosis process carried out manually by pathologists is quite time consuming and affects subjectivity. This study aims to develop a histopathology image-based breast cancer classification system using VGG-16. The dataset to be used consists of histopathology images that are grouped into 2 classes, namely benign and malignant. The data went through several preprocessing stages, including splitting and augmentation, to improve data quality. Test results show that this model achieves 91% accuracy, along with high precision, recall, and F1-scores on the test data. The performance of this model compares favorably with ensemble architectures such as, MobileNet, MobileNetV2. These findings indicate that the proposed approach can be an effective solution as a histopathology image-based breast cancer diagnosis tool.

Page 83 of 89 | Total Record : 889

 81   82   83   84   85 

Filter by Year

2019 2025


Reset
Filter By Issues
All Issue Vol 7 No 3 (2025): December 2025 Vol 7 No 2 (2025): September 2025 Vol 7 No 1 (2025): June (2025) Vol 6 No 4 (2025): March 2025 Vol 6 No 3 (2024): December 2024 Vol 6 No 2 (2024): September 2024 Vol 6 No 1 (2024): June 2024 Vol 5 No 4 (2024): March 2024 Vol 5 No 3 (2023): December 2023 Vol 5 No 2 (2023): September 2023 Vol 5 No 1 (2023): June 2023 Vol 4 No 4 (2023): March 2023 Vol 4 No 3 (2022): December 2022 Vol 4 No 2 (2022): September 2022 Vol 4 No 1 (2022): June 2022 Vol 3 No 4 (2022): March 2022 Vol 3 No 3 (2021): December 2021 Vol 3 No 2 (2021): September 2021 Vol 3 No 1 (2021): June 2021 Vol 2 No 2 (2020): December 2020 Vol 2 No 1 (2020): June 2020 Vol 1 No 2 (2019): December 2019 Vol 1 No 1 (2019): June 2019 Vol 1 No 1 (2019): Edisi Juni More Issue