Garuda - Garba Rujukan Digital

Securing Network Log Data Using Advance Encryption Standard Algorithm And Twofish With Common Event Format Ali, Moch. Dzikri Azhari; Hadiana, Asep Id; Melina, Melina
International Journal of Quantitative Research and Modeling Vol 5, No 3 (2024)
Publisher : Research Collaboration Community (RCC)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46336/ijqrm.v5i3.757

The rapid advancement of information technology demands enhanced security for data exchange in the digital world. Network security threats can arise from various sources, necessitating techniques to protect information transmitted between interconnected networks. Securing network logs is a critical step in strengthening overall network security. Network logs are records of activities within a computer network, including unauthorized access attempts, user activities, and other key events. This research focuses on developing a network log security system by comparing the performance of the Advanced Encryption Standard (AES) and Twofish algorithms, integrated with the Common Event Format (CEF) for encrypting network logs. Tests were conducted on network log datasets to evaluate system functionality and performance. Results indicate that the AES algorithm performs encryption and decryption faster than Twofish. Across five tests with different file sizes, AES took an average of 2.1386 seconds for encryption, while Twofish required 22.8372 seconds. For decryption, AES averaged 2.451 seconds compared to Twofish’s 26.140 seconds. The file sizes after encryption were similar for both algorithms. Regarding CPU usage, AES demonstrated higher efficiency. The average CPU usage during AES encryption was 0.5558%, whereas Twofish used 23.2904%. For decryption, AES consumed 0.4682% of CPU resources, while Twofish required 13.7598%. These findings confirm that AES is not only faster in both encryption and decryption but also more efficient in terms of CPU usage. This research provides valuable insights for optimizing network log security by integrating standardized log formats, like CEF, with appropriate encryption techniques, helping to safeguard against cyber threats.

Enhancing Email Client Security with HMAC and PGP Integration to Mitigate Cyberattack Risks Oktaviani, Ayu Nur; Hadiana, Asep Id; Melina, Melina
International Journal of Quantitative Research and Modeling Vol 5, No 3 (2024)
Publisher : Research Collaboration Community (RCC)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46336/ijqrm.v5i3.758

The rapid advancement of technology in the modern era has significantly increased the risk of data breaches and misuse, particularly in email communications. Ensuring data privacy and security is crucial to preventing information theft and mitigating cyberattack risks. This research focuses on enhancing email client security through the integration of Hash-Based Message Authentication Code (HMAC) and Pretty Good Privacy (PGP). HMAC is employed as a message authentication mechanism to ensure the integrity and authenticity of email messages, while PGP is utilized to generate public and private key pairs, enabling secure encryption and decryption processes. By integrating these two security methods into the email client system, we aim to enhance its resilience against cyber threats. The system's effectiveness was evaluated through black-box testing, demonstrating its capability to secure the email delivery process. Additionally, an analysis of key randomness using the entropy method revealed a maximum value of 6 bits, indicating a relatively high level of randomness and further strengthening the encryption process. The results of this study indicate that the combined use of HMAC and PGP provides a robust security solution for enhancing email client security and mitigating potential cyberattack risks.

CRYPTOCURRENCY TIME SERIES FORECASTING MODEL USING GRU ALGORITHM BASED ON MACHINE LEARNING Melina, Melina; Sukono, Sukono; Napitupulu, Herlina; Mohamed, Norizan; Herry Chrisnanto, Yulison; ID Hadiana, Asep; Kusumaningtyas, Valentina Adimurti
BAREKENG: Jurnal Ilmu Matematika dan Terapan Vol 19 No 2 (2025): BAREKENG: Journal of Mathematics and Its Application
Publisher : PATTIMURA UNIVERSITY

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30598/barekengvol19iss2pp1317-1328

The cryptocurrency market is experiencing rapid growth in the world. The high fluctuation and volatility of cryptocurrency prices and the complexity of non-linear relationships in data patterns attract investors and researchers who want to develop accurate cryptocurrency price forecasting models. This research aims to build a cryptocurrency forecasting model with a machine learning-based time series approach using the gated recurrent units (GRU) algorithm. The dataset used is historical Bitcoin closing price data from January 1, 2017, to July 31, 2024. Based on the gap in previous research, the selected model is only based on the accuracy value. In this study, the chosen model must fulfill two criteria: the best-fitting model based on the learning curve diagnosis and the model with the best accuracy value. The selected model is used to forecast the test data. Model selection with these two criteria has resulted in high accuracy in model performance. This research was highly accurate for all tested models with MAPE < 10%. The GRU 30-50 model is best tested with MAE = 867.2598, RMSE = 1330.427, and MAPE = 1.95%. Applying the sliding window technique makes the model accurate and fast in learning the pattern of time series data, resulting in a best-fitting model based on the learning curve diagnosis.

Klasifikasi Penyakit Stroke Menggunakan Metode Naïve Bayes Classification dan Chi-Square Feature Selection Benedictus Benny Sihotang; Yulison Herry Chrisnanto; Melina Melina
Voteteknika (Vocational Teknik Elektronika dan Informatika) Vol 12, No 3 (2024): Vol. 12, No 3, September 2024
Publisher : Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/voteteknika.v12i3.129022

Penyakit stroke merupakan suatu penyakit yang dapat memutuskan suplai darah menuju otak. Menurut World Health Organization (WHO), stroke merupakan salah satu penyebab kematian tertinggi di dunia. Penelitian ini bertujuan untuk menggunakan teknik klasifikasi untuk mendeteksi tingkat resiko terkena penyakit stroke. Klasifikasi merupakan teknik yang bertujuan untuk memperkirakan kelas dari suatu objek yang kelasnya masih tidak diketahui. Penelitian ini mengkombinasikan salah satu metode dari klasifikasi yaitu Naive Bayes dengan salah satu metode seleksi fitur yaitu Chi-Square untuk meningkatkan akurasi dari klasifikasi Naive Bayes. Hasil penelitian ini menunjukkan bahwa seleksi fitur Chi-Square terbukti dapat meningkatkan akurasi pada klasifikasi Naive Bayes dalam klasifikasi penyakit stroke dengan pembagian data latih dan uji yaitu 75:25. Hasil akurasi meningkat dari 73,55% sebelum menggunakan metode seleksi fitur Chi-Square menjadi 74,94% setelah menggunakan metode seleksi fitur Chi-Square. Penelitian ini diharapkan dapat membuka wawasan baru terkait metode seleksi fitur Chi-Square dalam meningkatkan kinerja dari suatu metode klasifikasi khususnya dalam mendeteksi risiko penyakit stroke sebagai tindakan pencegahan dan penanganan risiko penyakit stroke.Kata kunci: Chi-Square, klasifikasi,, Naive Bayes, Stroke. Stroke is a disease that can cut off the blood supply to the brain. According to the World Health Organization (WHO), stroke is one of the highest causes of death in the world. This study aims to use classification techniques to detect the risk level of stroke. Classification is a technique that aims to estimate the class of an object whose class is unknown. This research combines one of the classification methods, Naive Bayes, with one of the feature selection methods, Chi-Square, to improve the accuracy of Naive Bayes classification. The results of this study show that Chi-Square feature selection is proven to improve the accuracy of Naive Bayes classification on stroke disease classification with a division of training data and test data of 75:25. The accuracy results increased from 73.55% before using the Chi-Square feature selection method to 74.94% after using the Chi-Square feature selection method. This research is expected to open new insights related to the Chi-Square feature selection method in improving the performance of a classification method, especially in detecting the risk of stroke disease as a preventive measure and handling the risk of stroke disease.Keywords: Chi-Square, Classification, Naive Bayes, Stroke.

Peramalan Nilai Tukar Mata Uang Menggunakan Metode Nonlinear Autoregressive Exogenous Neural Network Cecep Jamaludin; Wina Witanti; Melina Melina
Voteteknika (Vocational Teknik Elektronika dan Informatika) Vol 12, No 3 (2024): Vol. 12, No 3, September 2024
Publisher : Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/voteteknika.v12i3.129132

Nilai tukar mata uang sering kali mengalami fluktuasi atau naik turun terhadap mata uang negara lain, terutama Great Britain Pound (GBP) terhadap Indonesian Rupiah (IDR). Perubahan nilai tukar mata uang dipengaruhi oleh berbagai faktor yang berhubungan langsung (endogen) dan faktor yang tidak berhubungan langsung (eksogen). Ketika nilai fluktuasi nilai tukar mata uang melebihi ambang batas tertentu dapat berdampak negatif pada perdagangan internasional dan menghambat pertumbuhan ekonomi negara. Penelitian ini mengkaji penggunaan metode Nonlinear Autoregressive Exogenous Neural Network (NARX-NN) untuk meramalkan nilai tukar mata uang GBP/IDR dengan menambahkan faktor eksternal seperti inflasi, suku bunga, ekspor, dan jumlah uang beredar pada model peramalan dengan tujuan untuk meningkatkan keakuratan peramalan nilai tukar mata uang dengan menggunakan metode NARX-NN. Hasil dari penelitian ini menunjukkan bahwa dengan memasukkan faktor-faktor yang mempengaruhi fluktuasi nilai tukar mata uang, diperoleh hasil peramalan yang lebih baik yaitu nilai Mean Absolute Error (MAE) sebesar 33.28, Root Mean Square Error (RMSE) sebesar 53.53, dan R-Squared ( sebesar 0.99 dengan pembagian data sebanyak 80/20. Diharapkan, hasil penelitian ini dapat menjadi referensi bagi investor, akademisi, dan masyarakat dalam memaksimalkan keuntungan dan meminimalisir risiko kerugian pada kurs mata uang GBP/IDR.Kata kunci : endogen; eksogen; nilai tukar; peramalan; NARX-NN. Currency exchange rates often fluctuate or rise and fall against other countries' currencies, especially the Great Britain Pound (GBP) against the Indonesian Rupiah (IDR). Changes in currency exchange rates are influenced by various factors that are directly related (endogenous) and factors that are not directly related (exogenous). When the value of currency exchange rate fluctuations exceeds a certain threshold, it hurts international trade and hampers the country's economic growth. This research examines the use of the Nonlinear Autoregressive Exogenous Neural Network (NARX-NN) method to forecast the GBP/IDR currency exchange rate by adding external factors such as inflation, interest rates, exports, and money supply to the forecasting model to improve the accuracy of forecasting currency exchange rates using the NARX-NN method. The results of this study show that by including factors that affect currency exchange rate fluctuations, better forecasting results are obtained, namely the Mean Absolute Error (MAE) value of 33.28, Root Mean Square Error (RMSE) of 53.53, and R-Squared (R2) of 0.99 with a data division of 80/20. It is hoped that this research can be a reference for investors, academics, and the public in maximizing profits and minimizing the risk of loss on the GBP/IDR currency exchange rate.Keywords: endogenous; exogenous; exchange rate; forecasting; NARX-NN.

Klasifikasi Kanker Payudara Berbasis Deep Learning Menggunakan Vision Transformer dengan Teknik Augmentasi Data Citra Ardiyansyah, Muhamad Salman; Umbara, Fajri Rakhmat; Melina, Melina
JURIKOM (Jurnal Riset Komputer) Vol 12, No 3 (2025): Juni 2025
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/jurikom.v12i3.8619

Breast cancer ranks among the leading causes of death in women worldwide. Early detection through mammographic image analysis plays a crucial role in increasing survival rates. However, manual interpretation of mammograms requires expert knowledge and is prone to errors. This study aims to develop a breast cancer classification model using mammography images based on the Vision Transformer (ViT) architecture without employing transfer learning. The dataset used is the Digital Database for Screening Mammography (DDSM), consisting of two categories: benign and malignant. To address class imbalance, undersampling and data augmentation techniques (flipping, rotation, cropping, and noise injection) were applied. All images were normalized and resized to 224×224 pixels to match the ViT input requirements. The model was trained for five epochs with a batch size of 16. Evaluation on the test data was conducted using seven metrics: accuracy, precision, recall, F1-score, Matthews Correlation Coefficient (MCC), Cohen’s Kappa Score, and Area Under the Curve (AUC). The results show that the model achieved an accuracy of 92.50%, precision of 90.48%, recall of 95.00%, F1-score of 92.68%, MCC of 85.11%, Kappa Score of 85.00%, and AUC of 95.75%. These findings indicate that the Vision Transformer is highly effective for mammographic image classification and holds potential as a reliable tool for automated breast cancer diagnosis support.

Prediksi Risiko Kesehatan Mental Mahasiswa Menggunakan Klasifikasi Naive Bayes Sumantri, Fithra Aditya; Chrisnanto, Yulison Herry; Melina, Melina
JURIKOM (Jurnal Riset Komputer) Vol 12, No 3 (2025): Juni 2025
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/jurikom.v12i3.8648

The mental health of university students is a growing concern as academic, emotional, and social pressures contribute to increased psychological risks. This study aims to classify mental health risk levels Low, Medium, and High among students using the Naïve Bayes classification algorithm. A dataset consisting of 1,000 entries and 11 key variables was utilized, covering academic, psychological, and behavioral factors. The preprocessing stage included data cleaning, label encoding, normalization, and rule-based labeling to determine the target classes. Model training and testing were conducted using stratified data splitting to preserve class distribution. The initial model achieved a classification accuracy of 88,67%, with macro average F1-score of 0.87 and weighted average F1-score of 0.88. Grid Search optimization with k-fold cross-validation was applied but showed no significant improvement, indicating the model was already in optimal configuration. Furthermore, probabilistic analysis revealed that Sleep Quality and Study Stress Level were the most influential features in predicting mental health risks. The findings suggest that Naïve Bayes is effective for multi-class classification with interpretable results. This research contributes to early detection efforts and offers a foundation for targeted interventions in university mental health management.

COMPARATIVE ANALYSIS OF TIME SERIES FORECASTING MODELS USING ARIMA AND NEURAL NETWORK AUTOREGRESSION METHODS Melina, Melina; Sukono, Sukono; Napitupulu, Herlina; Mohamed, Norizan; Chrisnanto, Yulison Herry; Hadiana, Asep ID; Kusumaningtyas, Valentina Adimurti; Nabilla, Ulya
BAREKENG: Jurnal Ilmu Matematika dan Terapan Vol 18 No 4 (2024): BAREKENG: Journal of Mathematics and Its Application
Publisher : PATTIMURA UNIVERSITY

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30598/barekengvol18iss4pp2563-2576

Gold price fluctuations have a significant impact because gold is a haven asset. When financial markets are volatile, investors tend to turn to safer instruments such as gold, so gold price forecasting becomes important in economic uncertainty. The novelty of this research is the comparative analysis of time series forecasting models using ARIMA and the NNAR methods to predict gold price movements specifically applied to gold price data with non-stationary and non-linear characteristics. The aim is to identify the strengths and limitations of ARIMA and NNAR on such data. ARIMA can only be applied to time series data that are already stationary or have been converted to stationary form through differentiation. However, ARIMA may struggle to capture complex non-linear patterns in non-stationary data. Instead, NNAR can handle non-stationary data more effectively by modeling the complex non-linear relationships between input and output variables. In the NNAR model, the lag values of the time series are used as input variables for the neural network. The dataset used is the closing price of gold with 1449 periods from January 2, 2018, to October 5, 2023. The augmented Dickey-Fuller test dataset obtained a p-value = 0.6746, meaning the data is not stationary. The ARIMA(1, 1, 1) model was selected as the gold price forecasting model and outperformed other candidate ARIMA models based on parameter identification and model diagnosis tests. Model performance is evaluated based on the RMSE and MAE values. In this study, the ARIMA(1, 1, 1) model obtained RMSE = 16.20431 and MAE = 11.13958. The NNAR(1, 10) model produces RMSE = 16.10002 and MAE = 11.09360. Based on the RMSE and MAE values, the NNAR(1, 10) model produces better accuracy than the ARIMA(1, 1, 1) model.

Klasifikasi Kanker Payudara Berbasis Deep Learning Menggunakan Vision Transformer dengan Teknik Augmentasi Data Citra Ardiyansyah, Muhamad Salman; Umbara, Fajri Rakhmat; Melina, Melina
JURNAL RISET KOMPUTER (JURIKOM) Vol. 12 No. 3 (2025): Juni 2025
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/jurikom.v12i3.8619

Breast cancer ranks among the leading causes of death in women worldwide. Early detection through mammographic image analysis plays a crucial role in increasing survival rates. However, manual interpretation of mammograms requires expert knowledge and is prone to errors. This study aims to develop a breast cancer classification model using mammography images based on the Vision Transformer (ViT) architecture without employing transfer learning. The dataset used is the Digital Database for Screening Mammography (DDSM), consisting of two categories: benign and malignant. To address class imbalance, undersampling and data augmentation techniques (flipping, rotation, cropping, and noise injection) were applied. All images were normalized and resized to 224×224 pixels to match the ViT input requirements. The model was trained for five epochs with a batch size of 16. Evaluation on the test data was conducted using seven metrics: accuracy, precision, recall, F1-score, Matthews Correlation Coefficient (MCC), Cohen’s Kappa Score, and Area Under the Curve (AUC). The results show that the model achieved an accuracy of 92.50%, precision of 90.48%, recall of 95.00%, F1-score of 92.68%, MCC of 85.11%, Kappa Score of 85.00%, and AUC of 95.75%. These findings indicate that the Vision Transformer is highly effective for mammographic image classification and holds potential as a reliable tool for automated breast cancer diagnosis support.

Prediksi Risiko Kesehatan Mental Mahasiswa Menggunakan Klasifikasi Naive Bayes Sumantri, Fithra Aditya; Chrisnanto, Yulison Herry; Melina, Melina
JURNAL RISET KOMPUTER (JURIKOM) Vol. 12 No. 3 (2025): Juni 2025
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/jurikom.v12i3.8648

The mental health of university students is a growing concern as academic, emotional, and social pressures contribute to increased psychological risks. This study aims to classify mental health risk levels Low, Medium, and High among students using the Naïve Bayes classification algorithm. A dataset consisting of 1,000 entries and 11 key variables was utilized, covering academic, psychological, and behavioral factors. The preprocessing stage included data cleaning, label encoding, normalization, and rule-based labeling to determine the target classes. Model training and testing were conducted using stratified data splitting to preserve class distribution. The initial model achieved a classification accuracy of 88,67%, with macro average F1-score of 0.87 and weighted average F1-score of 0.88. Grid Search optimization with k-fold cross-validation was applied but showed no significant improvement, indicating the model was already in optimal configuration. Furthermore, probabilistic analysis revealed that Sleep Quality and Study Stress Level were the most influential features in predicting mental health risks. The findings suggest that Naïve Bayes is effective for multi-class classification with interpretable results. This research contributes to early detection efforts and offers a foundation for targeted interventions in university mental health management.

Title

Found 34 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 34 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 34 Documents
Search