cover
Contact Name
Agus Tedyyana
Contact Email
agustedyyana@polbeng.ac.id
Phone
+6285289866666
Journal Mail Official
jurnaoinformatika@polbeng.ac.id
Editorial Address
Jl. Bathin alam, Sungai Alam Bengkalis-Riau 28711
Location
Kab. bengkalis,
Riau
INDONESIA
INOVTEK Polbeng - Seri Informatika
ISSN : 25279866     EISSN : -     DOI : https://doi.org/10.35314
Core Subject : Science,
The Journal of Innovation and Technology (INOVTEK Polbeng—Seri Informatika) is a distinguished publication hosted by the State Polytechnic of Bengkalis. Dedicated to advancing the field of informatics, this scientific research journal serves as a vital platform for academics, researchers, and practitioners to disseminate their insightful findings and theoretical developments. Scope and Focus: INOVTEK Polbeng - Seri Informatika focuses on a broad spectrum of topics within informatics, including but not limited to Web and Mobile Computing, Image Processing, Machine Learning, Artificial Intelligence (AI), Intelligent Systems, Information Systems, Databases, Decision Support Systems (DSS), IT Project Management, Geographic Information Systems, Information Technology, Computer Networks and Security, and Wireless Sensor Networks. By covering such a wide range of subjects, the journal ensures its relevance to a diverse readership interested in both the practical and theoretical aspects of informatics.
Articles 256 Documents
Evaluation of Multi-Algorithm Clustering for Marketplace MSME Segmentation Using a Big Data Analytics Approach Taufan, Anas; Redjeki, Sri
INOVTEK Polbeng - Seri Informatika Vol. 10 No. 3 (2025): November
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/k9gn7n75

Abstract

The rapid development of the digital economy has significantly driven MSME activity on marketplaces like Tokopedia, generating vast heterogeneous datasets. This study conducted a comparative evaluation of six clustering algorithms, including K-Means, Agglomerative Clustering, and GMM, using the Silhouette Score, Davies–Bouldin Index (DBI), and Calinski–Harabasz Index (CHI). Using a standardized Tokopedia MSME dataset from Yogyakarta, empirical results showed Silhouette scores ranging from 0.050 to 0.057, DBI from 0.45 to 0.53, and CHI from 950 to 1310. Although indicating low absolute cluster separation, these values facilitated meaningful relative comparisons. Among the tested algorithms, agglomerative clustering with Ward linkage demonstrated the best relative performance and consistency. Metric variability was examined through multiple runs to ensure stability. The analysis identified three segments: high-performing, medium-performing, and high-potential MSMEs, serving as a foundation for data-driven strategies. These findings underscore the necessity of a consistent multi-metric evaluation approach in MSME big data clustering studies.
An Indonesian Chatbot for Disease Diagnosis Using Retrieval-Augmented Generation Muhammad Adrinta Abdurrazzaq; Edwin Lesmana Tjiong; Aulia Fasya; Michelle Hiu; Joses Tanuwidjaya
INOVTEK Polbeng - Seri Informatika Vol. 10 No. 3 (2025): November
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/9nnkn955

Abstract

The rapid advancement of Large Language Models (LLMs) has enabled their use in medical information systems, although challenges such as hallucinations, domain mismatches, and the lack of a verified knowledge base remain significant, particularly in low-source languages ​​like Indonesian. This study introduces an Indonesian-language medical chatbot based on the open-source GPT-OSS-20B model enhanced through a Retrieval-Augmented Generation (RAG) pipeline. The system combines semantic retrieval using jina-embeddings-v3, lexical re-ranking with the BM25 algorithm, and a lightweight Logistic Regression-based domain filter as an initial filter to prevent out-of-domain LLM usage. Evaluation using Indonesian medical articles and annotated patient-doctor conversations shows that the domain filter works well on synthetic data but results in misclassification of natural queries. A hybrid weighted reranker (FAISS L2 + BM25) performed the best with a Top-30 accuracy of 0.699. Black-box testing indicates that the system flow functions as designed, although the response quality has not been validated by clinical experts. These findings suggest that RAG-based open-source LLMs can improve access to Indonesian-language medical information, but still have important limitations such as the lack of clinical validation, potential errors in scraped data, and suboptimal robustness of domain filters.
Early Warning and Real-Time Ship Tracking using AIS Data and Smartphone GPS Supria, Supria; Wahyat, Wahyat; Ryci Rahmatil Fiska, Ryci Rahmatil Fiska
INOVTEK Polbeng - Seri Informatika Vol. 10 No. 3 (2025): November
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/t62fpe35

Abstract

The high risk of maritime accidents in congested waters such as the Malacca Strait requires an affordable safety system specifically for small fishing vessels. This research proposes and evaluates a mobile-based early warning framework that integrates shore-based AIS data with fishermen’s smartphone GPS. The system was tested under 3 operational scenarios using 4G cellular networks over a coastal area of Bengkalis, involving 60 collision simulation events and 180 API requests. Performance evaluation shows an average system latency of 2.3 seconds with a maximum latency of 4.8 seconds. The early warning mechanism successfully detected dangerous proximity (≤50 meters) with an accuracy of 93.3% and an error rate of 6.7%. Position logging via JSON POST achieved a success rate of 96.1% during continuous operation for 2 hours. Although this study demonstrates improved situational awareness and reliable last-known position recording, the system currently uses distance-based detection and does not yet implement CPA/TCPA prediction, which remains future work. The framework contributes as a low-cost monitoring and early warning solution with potential support for SAR operations through reliable historical position data.
Residual-Gated Attention U-Net with Channel Recalibration for Polyp Segmentation in Colonoscopy Images Tanuwijaya, William; Yohannes
INOVTEK Polbeng - Seri Informatika Vol. 10 No. 3 (2025): November
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/4qmfa987

Abstract

This study proposed a modification to the Attention U-Net architecture by integrating a Residual-Gated mechanism and Squeeze-and-Excitation (SE) Block-based channel recalibration within the Attention Gate to enhance feature selectivity in polyp segmentation. This integration reinforces both spatial and channel attention, enabling the model to better highlight polyp regions while suppressing irrelevant background features. Experiments were conducted on three colonoscopy datasets, CVC-ClinicDB, CVC-ColonDB, and CVC-300, using IoU and DSC metrics. Compared to the Attention U-Net baseline, the proposed model achieves noticeable improvements, with performance gains of mIoU 0.0043 and mDSC 0.0094 on CVC-ClinicDB, mIoU 0.0012 on CVC-ColonDB, and a larger margin of mIoU 0.0224 and mDSC 0.0127 on CVC-300. The best results were obtained on CVC-ClinicDB (mIoU 0.8889, mDSC 0.9412). Although the absolute scores on CVC-ClinicDB and CVC-ColonDB are lower than those reported in several recent studies, these datasets contain higher variability in polyp size, boundary ambiguity, and illumination, contributing to more challenging segmentation conditions. Visual evaluation further shows smoother and more coherent boundaries, especially on small or low-contrast polyps. Overall, the integration of the residual-gated mechanism and SE block within the attention gate effectively improves model accuracy and generalization, particularly in challenging scenarios.
A Development of Worker Network Information System at the Batu City Manpower Using the Prototyping Method Okta; Suharso, Wildan
INOVTEK Polbeng - Seri Informatika Vol. 11 No. 1 (2026): February
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/vxt13a80

Abstract

The unemployment rate in Indonesia continues to pose significant social and economic challenges, requiring government initiatives such as job training programs. However, administrative processes that still rely on Google Forms lead to several issues, including data duplication (19 duplicates out of 598 registrations, or 3.2%), slow verification procedures (24–32 working hours per period), and limited real-time monitoring. This study focuses on developing a digital job training module integrated with the Worker Network Information System (SiJoker) at the Batu City Manpower Office. The system was developed using the prototyping method through three iterative cycles with direct user involvement, allowing the solution to be refined according to actual operational needs. The module includes participant registration, training management, and document validation features. System evaluation was conducted using Black Box Testing with 19 functional scenarios covering account management, training management, document management, verification, and reporting. The test results confirmed valid outputs for all scenarios without any critical errors. User evaluation by three staff members also validated system feasibility, particularly the effectiveness of explicit document status indicators, simplified navigation, and enhanced system responsiveness through optimized database queries.
Implementation of Concurrency Control to Prevent Race Condition in a Web-Based Billiard Table Reservation System Hermawan, Mohammad Luthfi; Suharso, Wildan
INOVTEK Polbeng - Seri Informatika Vol. 10 No. 3 (2025): November
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/vrnggg84

Abstract

The development of information technology has driven the digitalization of various services, including web-based billiard table reservation systems. However, web systems that operate in real time are prone to race conditions when multiple users attempt to book the same table simultaneously, potentially leading to double booking. This study aims to implement a Concurrency Control mechanism using the Firebase Transaction feature to prevent such booking conflicts. The research method adopts a Research and Development (R&D) approach with the ADDIE model, which consists of the stages of Analysis, Design, Development, Implementation, and Evaluation. Furthermore, testing was conducted through pre-test and post-test simulations across 10 trials with concurrent users ranging from 2 to 11 individuals. In the pre-test stage, all users were able to successfully book the same resource simultaneously, resulting in 100% double booking across all trials. In the post-test stage, after implementing the Concurrency Control mechanism using Firebase Transaction, only one request was accepted out of the same 10 trials, while all other requests were automatically rejected, resulting in 0% double booking. These findings demonstrate that the applied concurrency control mechanism is effective in maintaining data consistency and preventing race conditions in the web-based billiard table reservation system.
Optimization of Household Energy use Prediction Using Random Forest with Genetic Algorithm Feature Selection Nisrinaa Kamilah, Nyimas; Rahman, Abdul
INOVTEK Polbeng - Seri Informatika Vol. 10 No. 3 (2025): November
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/zedpkg51

Abstract

Electrical energy consumption continues to increase every year, so accurate prediction models are needed to support household electrical energy efficiency. This study analyzed high-resolution household electricity consumption data using the Random Forest (RF) algorithm and evaluated the influence of feature selection based on Genetic Algorithms (GA) in improving the performance of RF predictions. The base RF model achieves an RMSE of 0.6148, a MAE of 0.3478, and an R² of 0.5047. After implementing GA-based feature selection, the RF model with GA yields an RMSE of 0.6125 and an R² of 0.5084, indicating a marginal performance improvement. However, the MAE value increased slightly to 0.3503, which suggests that the increase was not uniform across the evaluation metrics. Overall, the RF approach with GA offers a modest improvement in prediction stability but with very limited accuracy gains, which highlights its potential and limitations in optimizing household energy consumption prediction.
Application of NFT and ERC-1155 Algorithm in Digital Certificate Verification Process Hardi, Azkaa Rahiila; Jaya, Safitri
INOVTEK Polbeng - Seri Informatika Vol. 11 No. 1 (2026): February
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/bezhpb82

Abstract

Digital certificate forgery remains a real problem in education and employment because traditional verification processes rely on centralized databases, are vulnerable to manipulation, and often take a long time. This study designs and implements a blockchain-based digital certificate verification system that models certificates as Non-Fungible Tokens (NFTs) using the ERC-1155 standard on the Manta Pacific Layer 2 network, and incorporates a Soulbound Token (SBT) mechanism to ensure that certificates cannot be transferred. The research adopts a prototyping method through eight stages, starting from architecture design and prototype development to the integration of ERC-1155 smart contracts with IPFS and wallets, as well as testing of minting functions, QR code-based verification, and rejection of asset transfers. The results demonstrate successful on-chain certificate issuance with significantly reduced transaction costs compared to ERC-72 based certificates on Layer 1 networks reported in previous studies, while maintaining a decentralized audit trail. The SBT implementation successfully rejects every attempt to transfer certificates to other wallets, thereby preventing the sale or illicit transfer of credential ownership. These findings indicate that the combination of ERC-1155, SBT, and IPFS on a Layer 2 network has strong potential as an efficient, secure, and practically adoptable digital certificate verification model for educational institutions.
Development of a Blockchain-Based LMS for Digital Learning Transparency Rahmat, Revo; Jaya, Safitri
INOVTEK Polbeng - Seri Informatika Vol. 11 No. 1 (2026): February
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/nyjn9270

Abstract

Current blockchain-based Learning Management Systems (LMS) are predominantly limited to partial certificate validation using ERC-721 and face significant scalability constraints and high transaction costs on Layer-1 infrastructures, failing to address the complete educational lifecycle. This study proposes a novel end-to-end decentralized LMS architecture integrating the Manta Pacific Layer 2 network for cost efficiency, the ERC-1155 standard for bulk license management, and Livepeer/IPFS protocols for autonomous content distribution. Employing a prototyping method, system performance was evaluated on the Manta Pacific Sepolia Testnet through 97 transaction scenarios covering course creation, enrollment, and real-time progress tracking via the Goldsky Indexer. Testing parameters focused on gas efficiency, transaction latency, and data integrity. Test results demonstrate significant operational efficiency with an average gas cost of 260.899 wei per transaction and a stable average block confirmation time of 10.0 seconds. Forensic validation confirmed 100% data consistency between internal system logs and blockchain explorer trails, alongside the successful execution of an automatic, intermediary-free revenue split (90/10). The proposed architecture proves capable of overcoming cost and latency barriers in educational blockchain adoption, offering a transparent, accountable, and technically feasible infrastructure for institutional scale.
Comparative Evaluation of Preprocessing Techniques in Twitter Sentiment Analysis for Indonesia’s 2024 Regional Elections Asro; Solihin
INOVTEK Polbeng - Seri Informatika Vol. 11 No. 1 (2026): February
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/tt65bb54

Abstract

The rapid expansion of social media has positioned Twitter as a critical platform for capturing public opinion during political events, including Indonesia’s 2024 Regional Elections. This study investigates the impact of preprocessing strategies and class balancing on the performance of sentiment analysis models applied to election-related tweets. An initial dataset of 9,096 tweets was collected and refined into 6,202 relevant entries from 2024–2025 through text cleaning, normalization, tokenization, and duplicate removal. Sentiment distribution analysis reveals a dominance of positive sentiment (58.4%), followed by negative (33.6%) and neutral (8.0%) expressions. Two classical machine learning classifiers—Naïve Bayes and Logistic Regression—were implemented using TF–IDF feature representation. To address class imbalance, the Synthetic Minority Oversampling Technique (SMOTE) was applied exclusively to the training data, and hyperparameter optimization was conducted using GridSearchCV. Model evaluation employed an 80/20 train–test split with accuracy, precision, recall, F1-score, and confusion matrices as performance metrics. Experimental results indicate that logistic regression combined with SMOTE and hyperparameter tuning achieved the highest accuracy of 93.08%, outperforming Naïve Bayes. The findings confirm that carefully designed preprocessing pipelines and class balancing significantly enhance the reliability of sentiment classification in political social media analysis.