Article Per Year (5 Year)
p-Index From 2020 - 2025
3.272
P-Index
This Author published in this journals
All Journal Jurnal Sistem Komputer & Kecerdasan Buatan JOURNAL OF INFORMATION SYSTEM RESEARCH (JOSH) Jurnal Teknik Informatika (JUTIF) International Journal of Engineering, Science and Information Technology Jurnal Pengabdian kepada Masyarakat Jurnal Pustaka AI : Pusat Akses Kajian Teknologi Artificial Intelligence Abdi Jurnal Publikasi Ranah Research : Journal of Multidisciplinary Research and Development Power Elektronik : Jurnal Orang Elektro Digitus : Journal of Computer Science Applications Jurnal Ilmu Komputer
Anggai, Sajarwo
Universitas Pamulang
Agriculture, Biological Sciences & Forestry Humanities Astronomy Biochemistry, Genetics & Molecular Biology Chemical Engineering, Chemistry & Bioengineering Chemistry Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Earth & Planetary Sciences Economics, Econometrics & Finance Education Electrical & Electronics Engineering Energy Engineering Environmental Science Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Library & Information Science Materials Science & Nanotechnology Mathematics Mechanical Engineering Physics Public Health Social Sciences Transportation Other

Published : 27 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 27 Documents
Search

 1   2   3 
Analisis Data Produksi Biskuit Dengan Algoritma Naive Bayes Dan Random Forest Sabarrudin; Agung Budi Susanto; Sajarwo Anggai
Jurnal Ilmu Komputer Vol 2 No 2 (2024): Jurnal Ilmu Komputer (Edisi Desember 2024)
Publisher : Universitas Pamulang

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

In the manufacturing industry, production problems often occur, often production does not match market demand, production is not well planned, therefore this study aims to develop a classification model using machine learning based on the Naive Bayes and Random Forest algorithms to classify biscuit production data. The main focus of this study is to utilize variables such as dough, number of mixers, production time parameters, and other relevant production factors to improve accuracy in classification. The dataset used in this study includes information from several previous production periods, namely data in 2019-2023, which is then used to train and test the Naive Bayes and Random Forest algorithm models. The training and validation process is carried out using commonly used model performance evaluation techniques. The results of the study show that the Random Forest model is able to provide high accuracy, namely 97.54% while Naive Bayes is 96.45%. Further analysis was also carried out to identify the variables that most influence production results, providing additional insights for optimizing the production process. The results of this study can contribute to the development of classification models for the food and beverage industry, especially in biscuit products, but also offer a more specific view of the factors that influence biscuit production. The implementation of this study can be a basis for manufacturers to make more precise and effective decisions in managing their production.
Analisis Sentimen Opini Masyarakat Terhadap Pemilu 2024 Melalui Media Sosial X Dengan Menggunakan Naive Bayes, K-Nearest Neighbor Dan Decision Tree Cut Shifa Khoirunnisa; Tukiyat; Sajarwo Anggai
Jurnal Ilmu Komputer Vol 2 No 2 (2024): Jurnal Ilmu Komputer (Edisi Desember 2024)
Publisher : Universitas Pamulang

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

This study aims to analyze public opinion sentiment towards the 2024 election using three machine learning classification algorithms: Naïve Bayes, K-Nearest Neighbors and Decision Tree. The data used in this study were taken from Social Media X, which is one of the social media platforms with a large and diverse data volume. The object of this study is public opinion expressed on Social Media X, with the subject of research in the form of tweets taken using the Twitter API, resulting in 5000 data with 2469 clean data. Data analysis involves text extraction and preprocessing processes that include data cleaning, tokenization, stopwords and stemming. The results of the study show the distribution of sentiment as follows: positive sentiment dominates with 96% of the total tweets, followed by neutral sentiment at 2% and negative sentiment at 1%. From the modeling results among the algorithms tested, K-Nearest Neighbors showed the best performance with an accuracy value reaching 97.50%, followed by Decision Tree having a performance with an accuracy value of 97.25% while Naïve Bayes had the lowest performance with an accuracy value of 96.14%. Although there is variation in performance among the algorithms used, none of them are completely consistent in classifying sentiment. This study makes a significant contribution in mapping public sentiment related to the 2024 election in Indonesia through data analysis from social media X, and provides insight into the effectiveness of various Data Mining Algorithms in sentiment analysis.
Analisis Dan Implementasi Sistem Manajemen Keamanan Informasi Menggunakan ISO/IEC 27001 (Studi Kasus Pada PT.XYZ) Wibowo, Rizki Septiyanto; Tukiyat; Sajarwo Anggai; Winarni
Jurnal Ilmu Komputer Vol 2 No 2 (2024): Jurnal Ilmu Komputer (Edisi Desember 2024)
Publisher : Universitas Pamulang

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

It has become a current necessity in every company regarding the implementation of information and communication technology governance in efforts to improve service quality. The implementation of information and communication technology governance is a critical factor in enhancing service quality across various companies. Therefore, the adoption of an Information Security Management System (ISMS) based on the ISO 27001:2013 standard becomes essential, in line with the conduct of regular audits to ensure its effectiveness. This research aims to develop and design an information security governance framework in accordance with ISO/IEC 27001 and to conduct audits on the system that has been implemented in PT. XYZ, to ensure its compliance with good and efficient standards. The methodology used is Plan-Do- Check-Act (PDCA), with data collection techniques through interviews and distribution of questionnaires for internal audits. The research findings indicate that the average ISO/IEC 27001 maturity level is at levels three and four. It is expected that this research can assist and provide recommendations related to security controlsused as guidelines and procedures for the implementation of information security, as well as ensuring the overall operation runs in accordance with ISO 27001 standards.
Optimizing Learning Rate, Epoch, and Batch Size in Deep Learning Models for Skin Disease Classification Rahman, Taufiqur; Anggai, Sajarwo; Arya Adhyaksa Waskita
Jurnal Ilmu Komputer Vol 3 No 1 (2025): Jurnal Ilmu Komputer (Edisi Juli 2025)
Publisher : Universitas Pamulang

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

This study explores the best combination of learning rate, number of epochs, and batch size for training deep learning models to classify skin diseases. The experiments involved analyzing how loss changes with learning rates on a logarithmic scale. The findings reveal that a learning rate of approximately 10-2 is most effective, with 5×10−3 offering additional stability during training. Various combinations of epochs and batch sizes were tested, ranging from 20 to 100 epochs and batch sizes between 32 and 128. The results show that using a batch size of 32 yielded the best outcomes, achieving a validation accuracy of 97.35% and the lowest validation loss of 0.1074. While a batch size of 128 was more efficient in terms of time, it resulted in slightly lower accuracy. The model performed optimally with 25 epochs and a batch size of 32, avoiding any signs of overfitting. Data preparation also played a crucial role, involving steps like image resizing, pixel normalization, and data augmentation to align with the requirements of models such as VGG-19, Inception-V4, and ResNet-152. Visualizing the dataset distribution ensured data quality and class balance, allowing the model to better recognize patterns. This study offers practical insights for effectively and efficiently training deep learning models, particularly for tasks related to skin disease classification.
Analisis Topik Penelitian Pendidikan Matematika Di Indonesia Dengan Menggunakan Metode Latent Dirichlet Allocation (LDA) junedi, Beni; Agung Budi Susanto; Sajarwo Anggai
Jurnal Ilmu Komputer Vol 3 No 1 (2025): Jurnal Ilmu Komputer (Edisi Juli 2025)
Publisher : Universitas Pamulang

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

On the research topic of Mathematics Education readers or researchers still have difficulty identifying research topics in the field of Mathematics Education. This is because there is no system or model that can be seen or used in determining research topics. Besides that, there is no automation of the research direction of Mathematics Education in Indonesia using topic modeling, so it is necessary to conduct a study or research on this. In research, the most important thing is the trend of research that is currently developing so that it can determine the novelty of the studies that have been done before. While there is no system used to determine trends and state of the art from research in the field of Mathematics Education. The aim of the research is to find out an overview of the research topics in Mathematics Education in Indonesia in 2020-2023 and to find out the implementation of modeling research topics in Mathematics Education in Indonesia using the Latent Dirichlet Allocation (LDA) method for 2020-2023. The research design consisted of literature study, data collection, data pre-processing: tokenization, case folding, stopword removal, and stemming, topic analysis with LDA, evaluation of the LDA method, and conclusions. Analysis of Topic Modeling with Latent Dirichlet Allocation using packages used from python including the Gensim and pyLDAvis packages. Based on the coherence score, the best number of topics (K) = 18, with a coherence score = 0.426 (the highest), it can be concluded that the number of topics produced is 18 topics.
Analisis Eksperimental Kinerja Transformers, VADER, dan Naive Bayes dalam Analisis Sentimen Teks Bahasa Indonesia: Studi Kasus Komentar Terkait Judi Online Sugiyo; Agung Budi Susanto; Sajarwo Anggai
Jurnal Ilmu Komputer Vol 3 No 1 (2025): Jurnal Ilmu Komputer (Edisi Juli 2025)
Publisher : Universitas Pamulang

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Sentiment analysis is a subfield of Natural Language Processing (NLP) that focuses on detecting and classifying opinions expressed in textual data. In the digital social context, the increasing volume of public comments related to online gambling in Indonesia highlights the need to map public perception. This study aims to conduct an experimental analysis of the performance of three popular sentiment analysis approaches: VADER (Valence Aware Dictionary and sEntiment Reasoner), Naive Bayes, and Transformers-based models, specifically on Indonesian-language text. The dataset consists of public comments from social media and digital platforms containing keywords related to online gambling. The research process involves text preprocessing, data labeling, model training (for Naive Bayes and Transformers), and performance testing. Evaluation metrics include accuracy, precision, recall, and F1-score. The experimental results show that the Transformers model (using IndoBERT) achieves the highest performance in terms of accuracy and generalization ability, while VADER performs less optimally due to its limitations in understanding Indonesian linguistic context. Naive Bayes demonstrates moderate and consistent performance but lacks the capability to capture complex contextual meanings. These findings contribute to selecting appropriate sentiment analysis methods for non-English languages and support the development of more accurate public opinion detection systems in the future
Prediksi Harga Cryptocurrency Menggunakan Algoritma Temporal Fusion Transformer, N-Beats dan Deepar Nugraha Wahyu, Fajar; Anggai, Sajarwo; Tukiyat, Tukiyat
Ranah Research : Journal of Multidisciplinary Research and Development Vol. 8 No. 1 (2025): Ranah Research : Journal Of Multidisciplinary Research and Development
Publisher : Dinasti Research

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.38035/rrj.v8i1.1949

Abstract

Cryptocurrency seperti Bitcoin, ETHereum, dan Solana memiliki volatilitas harga tinggi yang menyulitkan prediksi akurat. Penelitian ini bertujuan membandingkan akurasi tiga algoritma deep learning, yaitu Temporal Fusion Transformer (TFT), N-BEATS, dan DeepAR, dalam memprediksi harga harian ketiga aset tersebut. Data penelitian berupa harga penutupan, volume, dan kapitalisasi pasar yang diperoleh melalui CryptoDataDownload. Data diproses menggunakan normalisasi Min-Max Scaling, interpolasi linier untuk missing values, serta feature selection Pearson Correlation. Dataset kemudian dibagi ke dalam data pelatihan, validasi, dan pengujian dengan proporsi yang dapat disesuaikan, sehingga memungkinkan analisis pengaruh perbedaan pembagian data terhadap hasil model. Evaluasi dilakukan menggunakan MAE, RMSE, MAPE, dan R², serta uji statistik untuk menilai perbedaan signifikan antar model. Hasil penelitian menunjukkan bahwa N-BEATS memberikan performa terbaik dengan error paling rendah dan R² tertinggi, sementara TFT berada di urutan kedua dengan hasil yang cukup stabil. Sebaliknya, DeepAR secara konsisten memiliki performa terburuk dengan error tinggi dan R² negatif hampir di seluruh aset. Melalui eksperimen intensif, penelitian ini menunjukkan bahwa N-BEATS mengungguli TFT dan DeepAR dalam menjelaskan variansi data pada ketiga aset kripto: BTC, ETH, dan SOL. Pada semua dataset, N-BEATS mencapai nilai R² positif tertinggi di bawah Konfigurasi 2 (hidden size 32, 4 layers, dropout 0.3), dengan puncak 0.90 pada BTC, 0.93 pada ETH, dan 0.55 pada SOL. Nilai MAPE yang sesuai adalah 2.48% untuk BTC, 4.84% untuk ETH, dan 6.55% untuk SOL. Analisis juga mengungkap bahwa variasi ukuran hidden layer, epoch, dropout, jumlah layer, maupun pembagian data memengaruhi stabilitas serta performa prediksi, namun peningkatan kompleksitas tidak selalu menghasilkan performa yang lebih baik. Dengan demikian, N-BEATS dapat diidentifikasi sebagai model paling efektif untuk prediksi harga kripto, sekaligus memberikan kontribusi teoritis bagi pengembangan model peramalan deret waktu dan kontribusi praktis sebagai acuan bagi investor dalam pengambilan keputusan.
Co-Authors Abu Khalid Rivai Agnia Bilqisti AGUNG BUDI SUSANTO Agung Budi Susanto Ali Imron Arya Adhyaksa Waskita Brando, Charlo Chaerul Umam Cut Shifa Khoirunnisa Dhaman, Dhaman Effendy, Rangga Febrian Gumay, Rizki Izandi Hanto Sujatmiko Junedi, Beni Maulana, Romdon Muhamad, Faruk Muhammad Nana Muhammad Rafif Misbahuddin Murni Handayani Nugraha Wahyu, Fajar Pahira, Wulan Rafi Mahmud Zain Sabari Sabari Sabarrudin Slamet Wiyono Sugiyo Supriatna, Dahlan Sutrisno, Sutrisno T., Thoyyibah Taufiqur Rahman, Taufiqur Tukiyat Tukiyat Tukiyat, Tukiyat Waskita, Arya Adhyaksa Wibowo, Rizki Septiyanto Wibowo, Satria Ardi Wily, Wily Arisandi Winarni Zain, Rafi Mahmud