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 926 Documents
 76   77   78   79   80 
Stock Price Prediction Using LSTM and XGBoost with Social Media Sentiment Harani, Nisa Hanum; Marismati, Marismati
Building of Informatics, Technology and Science (BITS) Vol 7 No 2 (2025): September 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i2.8284

Abstract

The influence of social media on financial markets is growing and motivates research on the predictive role of sentiment in stock price movements. Bank Negara Indonesia (BBNI) is part of the Danantara holding company, and BBNI's strategic position is an important indicator for measuring the performance of the broader financial ecosystem in Indonesia. This study analyzes the influence of social media sentiment on the stock price prediction of Bank Negara Indonesia (BBNI), which is part of the state-owned holding company Danantara. Historical market data is combined with sentiment indicators obtained from public conversations on X/Twitter. Daily sentiment features are then integrated with market variables, including OHLCV data, to form a combined dataset. Two machine learning approaches were employed: Long Short-Term Memory (LSTM) and Extreme Gradient Boosting (XGBoost). The results revealed contrasting patterns between the two models. The LSTM Baseline consistently produced RMSE around (≈46–65) across all scenarios. However, XGBoost-Extended is the best-performing and recommended model for sentiment-integrated prediction with RMSE (≈30–40).
Narasi Presiden Indonesia: Analisis Wacana Politik Menggunakan BERTopic dalam Mengungkap Pola Tematik Pidato Presiden Uliyatunisa, Uliyatunisa; Tukiyat, Tukiyat; Waskita, Arya Adhyaksa; Handayani, Murni; Zain, Rafi Mahmud
Building of Informatics, Technology and Science (BITS) Vol 7 No 2 (2025): September 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i2.8298

Abstract

The speeches of the President of Indonesia play an important role as a means of political communication, policy delivery, and leadership image building in front of the public. However, the increasing volume of speeches presents new challenges in the manual analysis process, as it is time-consuming and prone to researcher subjectivity. This study offers a solution by using BERTopic, a transformer-based topic modelling method that utilises semantic representations from modern embedding models. The research data consists of transcripts of President Joko Widodo's official speeches obtained from the Cabinet Secretariat portal. To improve the quality of semantic representations, this study compares several Indonesian language embedding models, namely DistilBERT, NusaBERT, IndoE5, and SBERT. The analysis process was carried out through the stages of data preprocessing, embedding formation, dimension reduction, clustering, and model evaluation using topic coherence metrics. The objectives of this study were to reveal the themes contained in the President's speeches and to evaluate the effectiveness of embedding models in producing more coherent topics. The results show twenty main themes that consistently appear, including infrastructure development, economic policy, health and the pandemic, digital transformation, international diplomacy, sports, nationalism issues, and regional development. In terms of performance, SBERT provides the best results with a coherence value of UMass = -2.036 and NPMI = 0.082, indicating a positive semantic relationship. A UMass value close to zero indicates greater coherence of words within a topic, while an NPMI value above zero indicates that the connections between words are more easily understood by humans. This research contributes to the development of NLP-based political discourse studies in Indonesia, providing an empirical overview of the selection of appropriate embedding models in topic modelling and opening up opportunities for the integration of similar methods in public policy analysis.
Analisis Sentimen Publik Terhadap Deepfake AI Menggunakan Aplikasi X Dengan Metode Support Vector Machine dan Naive Bayes Classifier Al Afif, Satria; Suryono, Ryan Randy
Building of Informatics, Technology and Science (BITS) Vol 7 No 2 (2025): September 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i2.8303

Abstract

The rapid development of artificial intelligence (AI) technology has driven increased public interaction with AI-based platforms, including Deepfake AI. One of the main challenges that arises is how to objectively assess public opinion, particularly on social media, which serves as a primary medium for expressing opinions. This study aims to compare the performance of two machine learning algorithms, namely Support Vector Machine (SVM) and Naïve Bayes (NB), in analyzing public sentiment toward Deepfake AI on the X social media platform. The research dataset consists of 7,774 tweets collected between October and November 2024. After preprocessing, 5,559 tweets were used, categorized into three sentiment classes: positive, negative, and neutral. Data imbalance was addressed using the Synthetic Minority Over-sampling Technique (SMOTE), with 80% of the data allocated for training and 20% for testing. The results show that before applying SMOTE, the SVM algorithm achieved the highest accuracy at 71%, while Naïve Bayes only reached 62%. After the application of SMOTE, the performance of both algorithms improved, with SVM achieving 77% accuracy and Naïve Bayes reaching 68%. Thus, SVM proved to be the best-performing algorithm in this study, both before and after SMOTE application, delivering more balanced results across sentiment classes. This research demonstrates that sentiment analysis based on machine learning can be utilized to understand public opinion toward AI platforms, while also providing valuable insights for developers to improve service quality and strengthen public trust.
Integrasi K-Modes dalam Analisis Data Gizi Balita untuk Model Klasifikasi Risiko Stunting Desyanti, Desyanti; Renaldi, Reno; Mesran, Mesran
Building of Informatics, Technology and Science (BITS) Vol 7 No 2 (2025): September 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i2.8323

Abstract

The nutritional status of toddlers is an important indicator in assessing child growth and development and is closely related to the risk of stunting. However, the process of recording and classifying nutritional status at the Bukit Kapur Community Health Center is still done manually, making it prone to analysis delays and data processing errors. This study aims to implement the K-Modes algorithm in classifying toddler nutritional status based on categorical data, such as age, weight, and height. Toddler data were collected from the Bukit Kapur Community Health Center and underwent pre-processing, data transformation, and the application of the K-Modes algorithm to determine toddler nutritional groups. The results showed that the K-Modes algorithm was able to group toddler data into three main categories: well-nourished, at-risk of overnutrition, and overnourished. The majority of toddlers fell into the well-nourished category (98 toddlers), while only a small proportion fell into the at-risk of overnutrition and overnourished categories (1 toddler each). These findings indicate that the K-Modes method is effective in classifying toddler nutritional status based on categorical data and can assist health workers in monitoring child growth and preventing stunting.
Stacking Machine Learning Approach for Predicting Thermal Stability of Zinc–Metal Organic Frameworks (Zn-MOF) Pratama, Ananta Surya; Irnanda, Muhammad Diva; Umam, Taufiqul; Nugroho, Dandy Prasetyo; Azies, Harun Al
Building of Informatics, Technology and Science (BITS) Vol 7 No 2 (2025): September 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i2.8329

Abstract

Thermal stability is a fundamental parameter that determines the feasibility of Metal Organic Frameworks (MOF) for high-temperature industrial applications, including catalysis, gas purification, and energy storage. Experimental evaluation of thermal stability, while accurate, is often costly and time-consuming, highlighting the need for computational prediction models that are both efficient and dependable. This study develops a Quantitative Structure Property Relationship (QSPR) model using a stacking ensemble regression framework to predict the thermal stability of Zn-MOFs. The stacking approach combines Linear Regression, Lasso Regression, and Huber Regression as base learners, with Linear Regression serving as the meta-model, thereby leveraging the complementary strengths of individual algorithms. Results demonstrate that the stacking ensemble consistently outperformed all single models, delivering highly reliable predictions that remained stable across multiple validation scenarios. Furthermore, external validation with experimental data confirmed the model’s robustness and its ability to generalize beyond the training dataset. These findings underline the reliability of stacking as not only a tool for improving accuracy but also for ensuring predictive stability and reproducibility. The study highlights the potential of machine learning, particularly ensemble methods, as a powerful and trustworthy predictive framework for the rational design of thermally stable MOFs, offering both scientific and industrial significance in sustainable energy applications.
Comparison of Certainty Factor, Dempster Shafer, and Bayes' Theorem in Expert Systems for Diagnosing Female Reproductive System Diseases Mesran, Mesran; Rasli, Roznim Mohamad; Setiawansyah, Setiawansyah; Arshad, Muhammad Waqas
Building of Informatics, Technology and Science (BITS) Vol 7 No 2 (2025): September 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i2.8334

Abstract

Expert systems are one application of artificial intelligence used to mimic the ability of an expert in diagnosing a disease. This study aims to compare the performance of three inference methods Certainty Factor, Dempster-Shafer, and Bayes' Theorem in the diagnosis of female reproductive system diseases. Symptom data and expert knowledge values were obtained from medical experts to support the system's validity. Each method was implemented on the same symptom data, and the results were analyzed to assess the consistency of the diagnoses produced. The results show that the Certainty Factor method produced a diagnosis of Cervical Cancer with the highest confidence value of 0.9999, followed by the Dempster-Shafer method with the same diagnosis and a confidence value of 0.852. However, the Bayes Theorem method produced a different diagnosis, namely Ovarian Cyst, with a confidence value of 0.911. These differing results indicate that the characteristics and approaches of each method significantly influence the final diagnosis outcome. This study contributes insights to expert system developers regarding the strengths and weaknesses of each inference method. The selection of the appropriate method must be tailored to the system's requirements, data complexity, and the level of uncertainty in the medical information used.
Perbandingan Algoritma Naive Bayes, Random Forest, dan Support Vector Machine Terhadap Pandangan Masyarakat Mengenai Revisi Undang-Undang TNI di Instagram Nasrul, Royhan; Yudhistira, Aditia
Building of Informatics, Technology and Science (BITS) Vol 7 No 2 (2025): September 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i2.8164

Abstract

The revision of the Indonesian National Army Law (TNI Law), enacted in 2025, sparked widespread controversy within society, particularly concerning issues of civilian supremacy and potential military dominance. With the growing use of social media as a platform for public expression, platforms such as Instagram have become the primary medium for the public to voice their opinions regarding this issue. This study aims to analyze public sentiment toward the revision of the TNI Law by utilizing text classification algorithms, namely Naive Bayes, Random Forest, and Support Vector Machine (SVM). Data was collected from 28,669 Instagram comments and analyzed through stages of data crawling, preprocessing, and labeling. To address data imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) was applied. Subsequently, classification was performed using the three algorithms, with evaluation metrics including accuracy, precision, recall, and F1-score. The results after SMOTE demonstrated that the SVM algorithm delivered the best performance with an accuracy of (92%), followed by Random Forest at (88%), and Naive Bayes at (76%). Consequently, SVM was deemed the most effective in capturing patterns of public sentiment objectively. This research is expected to contribute to the advancement of digital public opinion studies and support the evaluation process of national defense policies
Klasifikasi Spam Bahasa Indonesia dengan IndoBERT dan XLM-RoBERTa: Evaluasi Pooling, Stride, dan Late-Fusion Darmono, Darmono; Saputro, Rujianto Eko; Barkah, Azhari Shouni
Building of Informatics, Technology and Science (BITS) Vol 7 No 2 (2025): September 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i2.8034

Abstract

Spam detection for Indonesian short messages such as SMS and email remains challenging due to lexical variation, character obfuscation, and class imbalance. This study provides a systematic evaluation to determine the most balanced configuration between accuracy and efficiency for Indonesian spam filtering. We compare two pretrained backbones (IndoBERT and XLM RoBERTa), along with representation strategies (truncation versus chunking), summarization schemes (pooling), and feature fusion approaches. The system follows a feature based design with an emphasis on simplicity, and is assessed using F1 Macro, spam class recall, AUPRC (Area Under the Precision Recall Curve), and efficiency metrics in terms of embedding build time and training latency. Results indicate that IndoBERT achieves superior binary classification performance with high efficiency, while XLM RoBERTa slightly outperforms on AUPRC, making it more suitable for risk ranking scenarios. Truncation combined with mean pooling consistently yields stable results. Although late fusion only provides marginal improvements, it remains relevant as it highlights the potential of domain specific signals to enhance robustness under heavy obfuscation. The final recommendation for production is IndoBERT with truncation, mean pooling, and embedding only. Limitations include the focus on short messages and the lack of evaluation under extreme obfuscation. Future work should explore character level augmentation, cross domain evaluation, and cost sensitive threshold tuning.
Komparasi Metode Naïve Bayes, Random Forest dan KNN untuk Analisis Sentimen Penambangan Nikel Setiyana, Beta Agus; Suryono, Ryan Randy
Building of Informatics, Technology and Science (BITS) Vol 7 No 2 (2025): September 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i2.8263

Abstract

The phenomenon of increasing natural resource exploitation in Indonesia’s conservation areas has raised significant public concern, one of which involves the planned nickel mining project in Raja Ampat, a region renowned for its extraordinary marine biodiversity. This plan has sparked debates between economic interests, environmental preservation, and the sociocultural values of local communities. Amid the growing public discourse, social media has become a major platform for people to express their opinions, support, or opposition toward mining activities. This study aims to map public sentiment regarding the nickel mining issue in Raja Ampat by analyzing 5,556 Indonesian-language tweets collected from the social media platform X using the keyword “save raja ampat” between January- June 2025. The data underwent several preprocessing stages, including cleaning, case folding, tokenizing, stopword removal, and normalization, and were then represented using the TF-IDF method. Sentiment labeling was performed semi automatically using a lexicon based approach into three categories: positive, neutral, and negative. The sentiment distribution showed dominance of neutral (72.9%), followed by negative (24.3%) and positive (2.8%), indicating class imbalance. To address this issue, the SMOTE technique was applied to the training data. Three classical algorithms K-Nearest Neighbor (KNN), Complement Naïve Bayes (CNB), and Random Forest (RF) were compared using cross-validation and holdout testing with accuracy, precision, recall, and F1-score as evaluation metrics. The results show that CNB performed most stably before SMOTE, while after SMOTE, KNN demonstrated significant improvement, especially in recall and macro F1-score. These findings confirm that the combination of data balancing techniques and classical algorithms remains relevant and efficient as a methodological baseline for public sentiment analysis on complex environmental issues such as nickel mining in Raja Ampat.
A Comparative Analysis of LSTM and GRU Models for AQI Forecasting in Tourist Destinations Ardianto, Luluk; Astuti, Yani Parti
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.v7i1.6633

Abstract

The Air Quality Index (AQI) is a critical metric for assessing air quality and its impact on human health, particularly in densely populated and tourist-heavy areas such as Malioboro, Yogyakarta. As one of Indonesia's most popular tourist destinations, the region experiences significant air quality fluctuations influenced by human activities, including transportation and tourism. This study evaluates the performance of two advanced deep learning models, Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU), in forecasting AQI and key pollutant parameters, PM10 and PM2.5, using two years of air quality data collected between January 2022 and December 2023. The results demonstrate that the LSTM model consistently outperforms GRU in predicting AQI (MSE: 163.757, RMSE: 12.797, MAE: 7.432, MAPE: 0.133) and PM2.5 (MSE: 32.001, RMSE: 5.657, MAE: 3.005, MAPE: 0.139), indicating its capability to model complex temporal patterns effectively. Conversely, the GRU model achieves better accuracy for PM10 predictions (MSE: 58.592, RMSE: 7.655, MAE: 4.168, MAPE: 0.180), showcasing its computational efficiency with competitive performance. These findings underscore the suitability of LSTM for applications prioritizing accuracy, while GRU provides a viable option for scenarios requiring faster computations. This research highlights the potential of leveraging deep learning models to tackle air quality challenges in urban and tourist areas, paving the way for informed decision-making and sustainable development initiatives

Page 78 of 93 | Total Record : 926

 76   77   78   79   80 

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