Garuda - Garba Rujukan Digital

Prediction of Five Elements Imbalance and Acupuncture Point Recommendations Using Health-LLM Agent Method for Symptom Diagnosis Based on Traditional Chinese Medicine (TCM) Theory at Acumastery Clinic Iwan Muttaqin; Arya Adhyaksa Waskita; Choirul Basir
International Journal Software Engineering and Computer Science (IJSECS) Vol. 5 No. 3 (2025): DECEMBER 2025
Publisher : Lembaga Komunitas Informasi Teknologi Aceh (KITA)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/ijsecs.v5i3.5775

Traditional Chinese Medicine (TCM) is a medical system that has been historically proven effective in diagnosing and managing various symptoms through the concepts of the Five Element imbalance, Yin-Yang, and acupuncture points. In the era of artificial intelligence, the utilization of Large Language Models (LLMs) specifically designed for the healthcare domain, referred to as Health-LLM Agents (AI-based health agents powered by LLMs), holds great potential in supporting TCM practices with greater efficiency and precision. This study aims to design and evaluate the performance of a Health-LLM Agent in predicting imbalances among the Five Elements (Wood, Fire, Earth, Metal, Water) based on patient symptoms, while also recommending appropriate acupuncture points for therapy. The methodology involves fine-tuning an LLM model with prompt engineering tailored to TCM terminology and principles, along with integrating symptom data in semi-structured text format. Evaluation is conducted using expert validation and classification metrics such as diagnostic accuracy, relevance of acupuncture point recommendations, and result interpretability. The findings indicate that the Health-LLM Agent achieves an 81% accuracy in predicting Five Element imbalances and receives 92% positive validation from TCM practitioners regarding acupuncture point recommendations. These results demonstrate that the Health-LLM Agent can serve as a promising tool to support the digitalization and personalization of TCM diagnosis through AI-based systems

ANALYSIS OF THE EFFECTIVENESS VALUE OF IMPLEMENTING THE TWO-TIER DECEPTION-DRIVEN SECURITY MODEL IN CYBER SECURITY SYSTEMS Sahrul Ramadhan; Agung Budi Sutanto; Arya Adhyaksa Waskita
INTECOMS: Journal of Information Technology and Computer Science Vol. 9 No. 2 (2026): INTECOMS: Journal of Information Technology and Computer Science
Publisher : Institut Penelitian Matematika, Komputer, Keperawatan, Pendidikan dan Ekonomi (IPM2KPE)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31539/h9n89079

The increasing complexity of cyber attacks, especially Brute Force and SQL Injection, poses a significant risk to production environments. Conventional reactive security measures are often unable to provide sufficient understanding regarding the behavior of attackers. This study designs and analyzes a "Two-Tier Deception Architecture" aimed at improving early warning capabilities without sacrificing the integrity of the production system. This architecture physically and logically separates the production environment as Tier 1 and the deception-based laboratory environment as Tier 2. By utilizing a combination of Fail2Ban and NFTables, the system stealthily redirects traffic from detected malicious actors to a separate environment hosting the Cowrie and DVWA honeypots. All security logs are collected and analyzed using a centralized ELK Stack SIEM. Evaluation using a curated dataset of 100 samples (consisting of 60 legitimate activities and 40 malicious activities) achieved a detection and redirection accuracy of 95%. The system demonstrates minimal resource usage on the production server while providing precise threat intelligence. This research shows that the inclusion of a deception tier within standard infrastructure substantially strengthens proactive defense and incident response effectiveness.

SIMULASI MITIGASI ZERO-TOUCH PADA SERANGAN BRUTE FORCE SSH DAN RDP BERBASIS ORKESTRASI SIEM WAZUH Putu Dedi Juliana; Arya Adhyaksa Waskita; Ferhat Aziz
INTECOMS: Journal of Information Technology and Computer Science Vol. 9 No. 2 (2026): INTECOMS: Journal of Information Technology and Computer Science
Publisher : Institut Penelitian Matematika, Komputer, Keperawatan, Pendidikan dan Ekonomi (IPM2KPE)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31539/w7g4ba46

Administrasi server berbasis Secure Shell (SSH, port 22) dan Remote Desktop Protocol (RDP, port 3389) pada infrastruktur layanan publik, khususnya Layanan Pengadaan Secara Elektronik (LPSE) Kabupaten Mahakam Ulu, secara inheren memperluas permukaan serangan terhadap Teknik brute force (MITRE ATT&CK T1110). Model mitigasi manual yang bergantung pada analis memperpanjang jarak antara deteksi kegagalan logon berulang dan kontainmen, sehingga membuka jendela eksploitasi yang dapat dimanfaatkan penyerang. Penelitian ini merancang dan memvalidasi prototipe simulasi mitigasi otomatis berbasis semantic Active Response Wazuh dalam kerangka Security Information and Event Management (SIEM). Arsitektur klien–pelayan (React/Vite pada sisi klien; Node.js/Express dengan persistensi JSON pada sisi pelayan) menjalankan tiga skenario pengujian fungsional: serangan brute force bersumber tunggal pada SSH, bersumber tunggal pada RDP, dan multi-sumber dengan tiga alamat IP berotasi. Setiap skenario menggunakan ambang 10 kegagalan autentikasi untuk memicu aturan deteksi 5710 (SSH) dan 60122 (RDP) beranotasi T1110, dilanjutkan eksekusi Active Response berupa firewall-drop (Linux) dan netsh.exe (Windows). Hasil pengujian menunjukkan Mean Time to Respond (MTTR) diskret sebesar satu tick simulasi pada ketiga skenario, dengan rasio keberhasilan isolasi alamat IP mencapai 100% terhadap himpunan sumber yang dimodelkan. Prototipe yang tervalidasi berfungsi sebagai cetak biru konseptual bagi perencanaan penerapan SIEM pada infrastruktur publik tanpa risiko gangguan layanan produksi.

Prediction of Five Elements Imbalance and Acupuncture Point Recommendations Using Health-LLM Agent Method for Symptom Diagnosis Based on Traditional Chinese Medicine (TCM) Theory at Acumastery Clinic Iwan Muttaqin; Arya Adhyaksa Waskita; Choirul Basir
International Journal Software Engineering and Computer Science (IJSECS) Vol. 5 No. 3 (2025): DECEMBER 2025
Publisher : Lembaga Komunitas Informasi Teknologi Aceh (KITA), Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/ijsecs.v5i3.5775

Traditional Chinese Medicine (TCM) is a medical system that has been historically proven effective in diagnosing and managing various symptoms through the concepts of the Five Element imbalance, Yin-Yang, and acupuncture points. In the era of artificial intelligence, the utilization of Large Language Models (LLMs) specifically designed for the healthcare domain, referred to as Health-LLM Agents (AI-based health agents powered by LLMs), holds great potential in supporting TCM practices with greater efficiency and precision. This study aims to design and evaluate the performance of a Health-LLM Agent in predicting imbalances among the Five Elements (Wood, Fire, Earth, Metal, Water) based on patient symptoms, while also recommending appropriate acupuncture points for therapy. The methodology involves fine-tuning an LLM model with prompt engineering tailored to TCM terminology and principles, along with integrating symptom data in semi-structured text format. Evaluation is conducted using expert validation and classification metrics such as diagnostic accuracy, relevance of acupuncture point recommendations, and result interpretability. The findings indicate that the Health-LLM Agent achieves an 81% accuracy in predicting Five Element imbalances and receives 92% positive validation from TCM practitioners regarding acupuncture point recommendations. These results demonstrate that the Health-LLM Agent can serve as a promising tool to support the digitalization and personalization of TCM diagnosis through AI-based systems

Spatiotemporal Dynamics of El Niño Modoki Impacts on East Java Rainfall Using EOF, BIRCH Clustering, and Wavelet Coherence Diah Ariefianty; Agung Budi Susanto; Arya Adhyaksa Waskita
Journal of Informatics and Vocational Education Vol. 9 No. 3 (2026): November 2026
Publisher : Informatics Education Department, Faculty of Teacher Training and Education, Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20961/joive.v9i3.3782

El Niño Modoki is a variation of El Niño characterized by sea-surface-temperature warming in the central Pacific, flanked by cooling in the eastern and western Pacific, and can influence rainfall patterns in Indonesia, particularly in East Java. This region was selected because it is one of Indonesia’s major food-producing areas, has high rainfall variability, and remains highly vulnerable to drought. This study aimed to analyze the impact of El Niño Modoki on monthly rainfall anomalies in East Java during 1991–2024. The data consisted of the El Niño Modoki Index (EMI) and monthly gridded rainfall data. The analytical methods included rainfall-anomaly calculation, Pearson correlation, Empirical Orthogonal Function (EOF), BIRCH clustering, Wavelet Coherence (WTC), and composite analysis. BIRCH clustering based on the first three EOF modes formed four rainfall-pattern clusters in East Java. WTC analysis showed that the relationship between EMI and rainfall was more dominant at interannual periods of approximately 1–4 years. Composite analysis indicated that El Niño Modoki reduced rainfall in East Java starting from the JJA period and became stronger and more spatially extensive during ASO. Overall, the impact of El Niño Modoki on East Java rainfall was spatial, seasonal, dynamic, and non-homogeneous across regions. These findings provide preliminary information for drought mitigation, water-resource management, and climate early-warning strengthening in East Java.

Analisis Sentimen Pengguna X Terhadap Pemilihan Gubenur Dki Jakarta Tahun 2024 Dengan Algoritma Naïve Bayes, K-Nearest Neighbor Dan Decision Tree Syarif Hidayatullah; Arya Adhyaksa Waskita; Achmad Hindasyah
IKRA-ITH Informatika : Jurnal Komputer dan Informatika Vol. 10 No. 1 (2026): IKRAITH-INFORMATIKA Vol 10 No 1 Maret 2026
Publisher : Fakultas Teknik Universitas Persada Indonesia YAI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37817/ikraith-informatika.v10i1.6283

The 2024 DKI Jakarta Gubernatorial Election is one of the main political agendas that attracts public attention in Indonesia. In this context, sentiment analysis of the Gubernatorial and Vice Gubernatorial candidates is important to understand public opinion. With the advancement of technology, especially the internet and social media platforms like X (formerly Twitter), the public can freely express their opinions regarding the candidates. This study aims to analyze public sentiment towards the three pairs of Gubernatorial and Vice Gubernatorial candidates of DKI Jakarta through tweets collected using the crawling method, resulting in 6,120 rows of data, with each candidate pair obtaining 2,040 rows of data. The accuracy of sentiment analysis was calculated using Naïve Bayes, K-Nearest Neighbor, and Decision Tree algorithms to compare the accuracy of these three algorithms with an 80:20 testing-training data split and feature extraction using TF-IDF and Transformer, implemented using the RapidMiner software. Based on the analysis results, Naïve Bayes with TF-IDF representation showed the highest accuracy for Paslon 1 at 86.76%, followed by Paslon 2 at 76.96%, and Paslon 3 at 72.79%. Meanwhile, K-NN with TF-IDF achieved the best results for Paslon 1 (67.40%) and Paslon 2 (71.32%), while Decision Tree achieved the highest accuracy for Paslon 1 at 72.55%. For the Transformer representation, the overall accuracy was lower compared to TF-IDF, with Paslon 1 achieving 56.86%, Paslon 2 at 53.43%, and Paslon 3 at 51.23%. These results indicate that TF-IDF is more effective for sentiment analysis of tweets related to the Gubernatorial candidates, with Naïve Bayes being the most accurate algorithm.

EDUKASI DAN SIMULASI PHISHING UNTUK MENINGKATKAN LITERASI KEAMANAN SIBER SISWA SMA SEJAHTERA 1 DEPOK Sapta Nugraha; Ferry Setefanus; Indriyani Astuti; Muhammad Al-Farisy; Ferdiansyah Winardi; Agung Sutopo; Abdul Haris; Arya Adhyaksa Waskita
Abdi Jurnal Publikasi Vol. 5 No. 1 (2026): September
Publisher : Abdi Jurnal Publikasi

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

High school students are active users of digital services, yet their ability to recognize phishing and social engineering attacks has not kept pace with that use. The partner school, SMA Sejahtera 1 Depok, did not yet have a structured, practice-based cybersecurity education program. This community service activity aimed to improve students' cybersecurity literacy and their ability to recognize and resist phishing and social engineering attacks. It was carried out as a one-day intensive workshop on 25 September 2026 for about 50 grade X students. The methods combined interactive lectures, case demonstrations, controlled simulations, role-play, and group discussions across four pillars: cybersecurity concepts, the anatomy of phishing, social engineering patterns, and account security practices. The case demonstration used a fake free food-and-drink e-voucher claim email impersonating a look-alike brand of a well-known coffee shop near the school. Evaluation used a pre-test and post-test, simulation worksheets, a phishing email simulation, and a satisfaction questionnaire. Of the 35 phishing emails delivered, 17 students (48.6%) opened the fake link and entered data into the imitation form without verification. The mean score rose from 90.67 on the pre-test (30 students) to 97.92 on the post-test (24 students), and correct identification of a safe address rose from 46.7% to 100%. These findings show that students remain vulnerable to familiar instant offers even when their knowledge is high, underlining the importance of experience-based education. All simulations were controlled, and the data entered by students were used only to count responses and were not published.

Classification of Low-Level Wind Shear on Imbalanced Radiosonde Data Using a Hybrid CNN-LSTM Architecture Sarnubih Hasan; Agung Budi Susanto; Arya Adhyaksa Waskita
CSRID (Computer Science Research and Its Development Journal) Vol 18 No 3 (2026): Oktober 2026
Publisher : LPPM Universitas Potensi Utama

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22303/csrid-.18.3.2026.426-438

Low-Level Wind Shear (LLWS) during the takeoff and landing phases poses a severe and dynamic threat to aviation safety. However, it remains inherently challenging to detect directly from surface observations due to its highly non-linear, transient, and sequential nature in the lower atmosphere. This study proposes an advanced deep learning approach to classify LLWS events based on high-resolution vertical radiosonde profile data spanning the 2024–2025 period at Soekarno-Hatta International Airport, Indonesia. Addressing the critical challenge of a highly imbalanced dataset—where hazardous meteorological events are exceptionally rare compared to normal atmospheric conditions—the classification target was simplified into an event-level binary scheme. This scheme utilizes a hazard intensity threshold of ≥ 4 knots/100 ft, adhering strictly to the International Civil Aviation Organization (ICAO) moderate-to-severe hazard categories. The comprehensive research methodology involved fixed-grid interpolation into 21 uniform vertical observation points (spanning 0 to 2000 ft), rigorous wind vector transformation into zonal and meridional components, and circular encoding for wind direction to maintain absolute numerical stability during model training. A hybrid Convolutional Neural Network and Long Short-Term Memory (CNN-LSTM) architecture was developed to capture both local spatial gradients and long-term vertical dependencies, which was then systematically compared against a Transformer baseline equipped with a global self-attention mechanism. The empirical results and operational discussion demonstrate that the proposed CNN-LSTM model significantly outperformed the Transformer baseline, achieving an outstanding Area Under the Precision-Recall Curve (PR-AUC) of 0.9672 compared to the Transformer's 0.7533. Furthermore, the confusion matrix analysis conclusively proved the superior capability of the CNN-LSTM network in minimizing false negatives (missed detections) from 17 down to merely 7 critical events. In conclusion, the integration of local feature extraction through CNN and sequential dependency modeling via LSTM is highly effective for addressing imbalanced meteorological datasets, providing a highly robust and physically consistent computational foundation for automated aviation safety early warning systems.

Perfomance Comparison of LSTM and Transformer Model for Rainfall Data Imputation on Automatic Weather Station Tri Nurmayati; Agung Budi Susanto; Arya Adhyaksa Waskita
CSRID (Computer Science Research and Its Development Journal) Vol 18 No 3 (2026): Oktober 2026
Publisher : LPPM Universitas Potensi Utama

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22303/csrid-.18.3.2026.491-501

High-resolution rainfall observations from Automatic Weather Stations (AWS) are essential for hydrometeorological analysis, yet operational sensor and communication failures frequently create missing values. This study compared Long Short-Term Memory (LSTM) and Transformer architectures for imputing missing 10-minute rainfall observations recorded at the Banten Climatological Station AWS in South Tangerang during 2021-2025. Rainfall was modeled using a sliding window containing rainfall at the previous three intervals together with concurrent air temperature and relative humidity. The chronologically ordered dataset comprised 242,497 valid samples and was divided into training, validation, and testing subsets using a 60:20:20 ratio. Each architecture was trained with the Adam optimizer for up to 100 epochs under batch sizes of 32, 16, and 8, with early stopping used to control overfitting. Performance was assessed using root mean square error, mean absolute error, and the coefficient of determination. The best LSTM configuration used a batch size of 8 and achieved an RMSE of 1.1520, an MAE of 0.5140, and an R² of 0.6626 on the test set. The Transformer with a batch size of 8 obtained an RMSE of 0.8287, an MAE of 0.3635, and an R² of 0.8244. These results demonstrate that self-attention represented short-range multivariate dependencies more effectively than recurrent memory under the imbalanced rainfall distribution Keywords :automatic weather station; imputation; rainfall; LSTM; tansformer..

Intelligent DevSecOps Automation Framework Based on AI-Driven Security Analytics for Agent-Based Jenkins Pipelines Risko; Agung Budi Susanto; Arya Adhyaksa Waskita
Jurnal Teknologi Informatika dan Komputer Vol. 12 No. 2 (2026): Jurnal Teknologi Informatika dan Komputer
Publisher : Universitas Mohammad Husni Thamrin

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37012/jtik.v12i2.3615

The adoption of Continuous Integration/Continuous Delivery (CI/CD) has accelerated the implementation of DevSecOps as an approach for integrating security throughout the software development lifecycle. Although various automated security tools are widely available, their outputs remain fragmented, require extensive manual analysis, and provide limited support for security-driven decision-making within CI/CD pipelines. This gap creates the objective of this study, which is to design and evaluate an Intelligent DevSecOps Automation Framework that combines AI-Driven Security Analytics with an Agent-Based Jenkins Pipeline to integrate multiple security tools, normalize heterogeneous findings, enforce a unified security quality gate, and generate contextual security advisories through Large Language Models (LLMs). The study employed an applied research method with a prototype development and experimental approach. The proposed framework integrates Ruff, Semgrep, SonarQube, and Trivy, and is supported by a finding normalization and consolidation mechanism, risk-based prioritization, security quality gate evaluation, and AI-driven advisory generation orchestrated through LangChain. The framework was evaluated using a deliberately vulnerable payment-service repository to examine its capability in detecting vulnerabilities, consolidating cross-tool findings, classifying risk, and producing mitigation recommendations together with a release decision. The results show that the framework successfully unified four security tools within a single pipeline, producing 160 raw findings that were normalized and consolidated into 37 unique findings, of which 27 were categorized as Critical or Major. Fifteen of the highest-priority findings were forwarded to the AI-Driven Security Analytics component, yielding an Overall Risk and Exploitation Risk of Critical, 14 items classified as release blockers, and an overall release recommendation of Hold. These findings indicate that the proposed framework effectively supports early vulnerability detection, cross-tool consolidation, risk-based prioritization, security quality gate enforcement, and AI-assisted security advisory generation to strengthen security decision-making in modern DevSecOps pipelines.

Title

Found 11 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 11 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 11 Documents
Search