cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
-
Contact Email
-
Phone
-
Journal Mail Official
-
Editorial Address
-
Location
Kota medan,
Sumatera utara
INDONESIA
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING
Published by Universitas Medan Area
ISSN : 25496247     EISSN : 25496255     DOI : -
Core Subject : Science, Engineering,
Computer Science & IT Engineering
JURNAL TEKNIK INFORMATIKA, JITE (Journal of Informatics and Telecommunication Engineering) is a journal that contains articles / publications and research results of scientific work related to the field of science of Informatics Engineering such as Software Engineering, Database, Data Mining, Network, Telecommunication and Artificial Intelligence which published and managed by the Faculty of Informatics Engineering at the University of Medan Area .
Arjuna Subject : -
Articles 464 Documents
 43   44   45   46   47 
Sentiment Analysis of Action Mobile Application Reviews Using Logistic Regression and Support Vector Machine Haniful Fikri; Nurdin; Nunsina
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Vol. 10 No. 1 (2026): Issues July 2026
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jite.v10i1.18024

Abstract

The digital transformation has driven PT. Bank Aceh Syariah to launch the Action Mobile application. Despite its benefits for customers, user reviews on the Google Play Store indicate varying perceptions due to differences in digital experiences. This study aims to analyze user sentiment toward the Action Mobile application while comparing the effectiveness of the Logistic Regression and Support Vector Machine (SVM) algorithms. A total of 3,000 clean review data were collected through web scraping techniques. The dataset exhibits an imbalanced distribution, dominated by 1,840 positive reviews (61.33%), followed by 821 negative reviews (27.37%), and 339 neutral reviews (11.30%). Model testing was conducted using the 10-Fold Cross Validation so that each data has the opportunity to become test data and the evaluation results become more objective, utilizing TF-IDF for word weighting. The evaluation results using a 3 × 3 multiclass confusion matrix based on a weighted average demonstrate that the Logistic Regression algorithm outperforms SVM across all testing metrics. The Logistic Regression model successfully achieved an Accuracy of 0.9023, Precision of 0.897, Recall of 0.9023,, and an F1-Score of 0.895. Meanwhile, the SVM model obtained an Accuracy of 0.9013, Precision of 0.8969, Recall of 0.9013, and an F1-Score of 0.898. This performance variance proves that the Logistic Regression architecture is more adaptive and optimal for this specific case study. The findings of this study are expected to serve as evaluation material for enhancing Action Mobile services.
Comparison of Naïve Bayes and Logistic Regression for MBG Sentiment Analysis on X Fibry Widianti; Nurdin; Fajriana
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Vol. 10 No. 1 (2026): Issues July 2026
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jite.v10i1.18025

Abstract

The Free Nutritious Meal Program (MBG) is a strategic step by the Indonesian government to improve the quality of human resources through nutritional improvement. However, this policy has triggered polarization of opinion on social media X, ranging from support to criticism regarding distribution and transparency. This study aims to analyze public perception of the MBG program by comparing the performance of the Naïve Bayes and Logistic Regression algorithms. A total of 1,500 tweets were collected through scraping techniques using Tweetharvest. The research stages included text preprocessing and feature extraction using TF-IDF, with a 70:30 split between training and testing data. The evaluation results showed that Logistic Regression had superior performance with an accuracy of 72%, a precision of 72.17%, a recall of 69.32%, and an f1-score of 69.60%. Meanwhile, Naïve Bayes achieved an accuracy of 70.22%, a precision of 80.52%, a recall of 65.36%, and an f1-score of 63.70%. Research findings indicate that Logistic Regression is more dominant in predicting negative and neutral sentiment, while Naïve Bayes tends to dominate predictions of positive sentiment. Overall, Logistic Regression has proven more consistent and effective in mapping the dynamics of public perception of government policies on digital platforms than Naïve Bayes.
Klasifikasi Risiko Penyakit Kardiovaskular Menggunakan Metode Stacking Classifier Dan Explainable AI M. Hanif Mulyazam; Jondri Jondri; Indwiarti
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Vol. 10 No. 1 (2026): Issues July 2026
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jite.v10i1.18051

Abstract

Penyakit kardiovaskular secara konsisten menjadi penyebab utama kematian secara global. Keterlambatan penanganan medis sering terjadi karena keterbatasan metode konvensional dalam mendeteksi pola risiko klinis secara dini. Penelitian ini membangun model Stacking Classifier dengan Random Forest, XGBoost, Support Vector Machine, dan Multi-Layer Perceptron sebagai base learner, serta Logistic Regression sebagai meta-learner. Model dilatih menggunakan 70.000 observasi publik dan menghasilkan 60.383 baris data bersih setelah penghapusan data ekstrem, data dengan kombinasi input identik tetapi label target berbeda, serta observasi dengan tekanan darah yang tidak logis. Evaluasi dilakukan menggunakan Stratified K-Fold 5-fold. Hasil rata-rata validasi silang menunjukkan akurasi 73,94 persen, recall 72,24 persen, F1-score 73,96 persen, dan ROC AUC 0,8056. Fold terbaik diperoleh pada fold kedua dengan ambang batas 0,4638, akurasi 74,00 persen, recall 76,27 persen, F1-score 75,06 persen, dan ROC AUC 0,8066. Confusion Matrix pada fold terbaik menunjukkan 4.724 True Positive dan 1.470 False Negative, sehingga model lebih tepat diposisikan sebagai pendukung penapisan awal, bukan sebagai alat diagnosis akhir. Analisis SHapley Additive exPlanations (SHAP) pada model Stacking Classifier dari fold terbaik menunjukkan bahwa tekanan darah sistolik, usia, kolesterol, dan tekanan darah diastolik merupakan fitur yang paling dominan dalam pembentukan prediksi risiko.
Optimization of HIV/AIDS Classification Using the SMOTE Technique and CatBoost Algorithm Annisa Fadhillah Pulungan; Chairil Umri; Rossy Nurhasanah
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Vol. 10 No. 1 (2026): Issues July 2026
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jite.v10i1.18134

Abstract

Despite various mitigation efforts, Human Immunodeficiency Virus (HIV)/Acquired Immunodeficiency Syndrome (AIDS) remains a significant public health issue with widespread impacts in Indonesia. One of the challenges in HIV/AIDS classification using machine learning is data imbalance, where the number of HIV cases is smaller than Non-HIV cases. The aim of this study is to analyze the performance of the CatBoost algorithm in classification tasks and to evaluate the impact of the Synthetic Minority Oversampling Technique (SMOTE) on improving model performance in imbalanced datasets. The research method involves applying the CatBoost algorithm to the original dataset as well as to data that has been processed using SMOTE-based oversampling. Furthermore, model performance is evaluated using Precision, Recall, F1-Score, and Precision-Recall Area Under Curve (PR-AUC) metrics. The SMOTE + CatBoost model achieved an accuracy of 95%, precision of 93%, recall of 92%, F1-Score of 93%, and PR-AUC of 0.953, all of which are higher than those of the CatBoost Baseline model. In addition, the number of undetected HIV cases was reduced from 28 to 13 cases. The findings indicate that the integration of SMOTE with the CatBoost algorithm improves model performance, resulting in better classification outcomes on imbalanced datasets compared to the CatBoost Baseline, and potentially supports a more effective HIV/AIDS early detection system.
Real-Time Lightweight Monitoring System for 5G Standalone Networks Using MQTT and Node-RED Muhammad Juan Farza Rafly Alganiyu; Sopian Soim; Mohammad Fadhli
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Vol. 10 No. 1 (2026): Issues July 2026
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jite.v10i1.18144

Abstract

The rapid development of 5G Standalone networks requires a monitoring system capable of operating in real time while maintaining efficient computational resource utilization. This study aims to design and implement a lightweight monitoring system based on Message Queuing Telemetry Transport and Node-RED for a Universal Software Radio Peripheral B210-based 5G Standalone infrastructure. The proposed system applies a publish/subscribe communication architecture in which telemetry data are distributed through Message Queuing Telemetry Transport and visualized using a Node-RED dashboard interface. The research environment utilizes Open5GS and srsRAN to generate and monitor network parameters in real time. The results demonstrate that the developed monitoring system is capable of displaying telemetry information responsively with stable data transmission performance and low computational overhead. In addition, the lightweight telemetry architecture successfully supports centralized monitoring without significantly affecting the operational performance of the main system. Therefore, the proposed approach can be considered an effective alternative solution for supporting efficient and real-time operational monitoring in Software Defined Radio-based 5G Standalone environments
Analysis of 5G Standalone Security Against Man-in-the-Middle Attacks Using Sequence Number Desynchronization An Nisa; Sopian soim; Lindawati
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Vol. 10 No. 1 (2026): Issues July 2026
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jite.v10i1.18198

Abstract

The architecture of standalone 5G networks offers improved communication performance, but its authentication procedures remain vulnerable to eavesdropping threats. This research evaluates the system’s vulnerability to Man-in-the-Middle attacks and proposes comprehensive mitigation methods to address these security issues. Relying on Open5GS as the core network and UERANSIM as a simulator for base stations and user devices, testing was conducted in a virtualized environment. The methods used include traffic manipulation at the N2 interface to trigger desynchronization of the sequence number in the 5G-AKA authentication procedure. Experimental results show that packet interception successfully caused sequence number desynchronization, which directly led to authentication failure and device isolation due to Denial of Service. Additionally, network monitoring revealed that the HTTP/2 protocol displays customer data in plain text. In conclusion, Secure Socket Layer (SSL) encryption has proven effective as a mitigation method. With this integration, internal communication channels have been converted to a fully encrypted HTTPS format, preventing eavesdropping attempts. Furthermore, customer cryptographic parameters remain protected while keeping the operational overhead below acceptable limits.
Web-Based RESTful Monitoring of Water Level and Nutrient Concentration In Hydroponic Systems Rahmat Fadli Isnanto; M. Dzaki Akhyar; Sutarno; Sarmayanta Sembiring
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Vol. 10 No. 1 (2026): Issues July 2026
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jite.v10i1.18208

Abstract

Hydroponics requires continuous monitoring of water level and nutrient concentration. This study implements a web-based monitoring system with ESP32, TDS sensor, and water level sensor. Sensor data is sent via RESTful API to a Laravel server and displayed on a responsive dashboard. The main scientific contribution is the implementation of a decoupled RESTful API middleware between hardware and presentation layers, differing from platform-centric approaches. Analysis of 41 sensor readings revealed two operational clusters. The first cluster (20 samples) showed an average of 279.5 ppm (CV 15.2%) with constant water level at 4.8 cm. The second cluster (21 samples) showed an average of 3450.2 ppm (CV 3.5%) with water level at 6.4 cm. Two outliers (3963 ppm and 3751.6 ppm) were identified in the high-concentration cluster, likely due to transient mixing. No significant temporal trends were observed within clusters. API testing achieved 100% success rate, latency <750 ms, and theoretically supports throughput of 185 requests/second with horizontal scalability. A traceability matrix confirmed all six functional requirements. The decoupled RESTful middleware overcomes IoT platform interoperability limitations, enabling future expansion and supporting real-time hydroponic monitoring for urban agriculture
Evaluation of the Impact of Online Monitoring on Lead Time in an Experimental Assembly Line Esa Belia; Akbar Maulana Yusuf; Mohammad Fauzan Aulialdi; Muhammad Arifudin Lukmana
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Vol. 10 No. 1 (2026): Issues July 2026
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jite.v10i1.18364

Abstract

Advances in modern manufacturing require real-time visibility into production data to avoid delayed decision-making and inventory buildup; however, current monitoring technologies are expensive and complex for small industries. This study was conducted on an experimental assembly line by designing, implementing, and evaluating an IoT-based online monitoring system. The experiment compared the 4-station toy airplane assembly process before and after system implementation, with time data validated using the adequacy test, Shapiro-Wilk normality test, and paired t-test. The analysis results showed a significant improvement in production flow performance, with a reduction in lead time at each workstation of 25,21%, 22,37%, 21,91%, dan 30,35%., respectively. With real-time data transmission to a cloud database, early detection of bottlenecks through work-in-process monitoring can be performed, which subsequently improves the balance of the assembly line.
A Comparative Analysis of ResNet50, ConvNeXtTiny, and Vision Transformer for Rice Leaf Disease Classification susilawati susilawati; Stephanus Priyowidodo; Andre Hasudungan Lubis
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Vol. 10 No. 1 (2026): Issues July 2026
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jite.v10i1.18397

Abstract

Rice leaf diseases can significantly reduce crop productivity and grain quality, making rapid, objective, and accurate identification essential. This study compares the performance of three deep learning architectures, namely ResNet50, ConvNeXtTiny, and Vision Transformer (ViT-B16), for rice leaf disease classification. The dataset consists of 5,932 images representing four disease classes: Bacterial Blight, Blast, Brown Spot, and Tungro. The data were divided into 3,559 training images, 1,186 validation images, and 1,186 testing images using the same random seed. All models were trained using transfer learning with ImageNet pretrained weights, data augmentation applied only to the training set, dropout regularization, and a two-stage training strategy involving feature extraction and fine-tuning. Performance was evaluated using validation loss, accuracy, precision, recall, F1-score, training curves, and confusion matrices. The results show that ResNet50 achieved the best performance, with a validation accuracy of 99.92% and a validation loss of 0.00326, followed by ViT-B16 (99.16%) and ConvNeXtTiny (98.99%). All models classified the Tungro class with high accuracy, while minor misclassifications occurred among the Bacterial Blight, Blast, and Brown Spot classes due to their similar visual characteristics. Overall, ResNet50 proved to be the most effective model for rice leaf disease classification, although validation using real-world field images is still required before practical deployment
Antenna Azimuth and Radius Optimization in Heterogeneous Cellular Networks Based on Genetic Algorithms Mulyono Mulyono; Oktaf Brillian Kharisma; Hasdi Radiles
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Vol. 10 No. 1 (2026): Issues July 2026
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jite.v10i1.18416

Abstract

As cellular networks evolve into microcells or smaller ones, the number of sites required to serve an area will also increase, making network optimization impossible to perform manually. This study provides an overview of the implementation of Genetic Algorithms in solving radius and azimuth configuration problems for cell coverage. The main contribution of this study is the use of Fixed Step and Variable Step azimuth selection methods in the Genetic Algorithm coding process. This study also raises the issue of interference detection mechanisms in two adjacent cells. The problem presentation method uses a model with a measurable level of complexity by deriving from a homogeneously distributed layout model. This study uses Matlab in building an algorithm simulator and MapInfo in displaying the results visually. The results show that the proposed Variable Step method provides better optimization performance than the Fixed Step method, which was only able to resolve 62.5% of interference spots. To account for the stochastic nature of the Genetic Algorithm, the simulation was executed 10 independent times using identical parameter settings. Statistical evaluation indicates that the algorithm achieved an average fitness value of 59.40 with a standard deviation of 3.84, while requiring an average computational time of 635.41 seconds with a standard deviation of 9.23 seconds. Furthermore, the algorithm reached convergence at an average generation of 957.10 with a standard deviation of 34.98 generations. These results demonstrate that the proposed algorithm exhibits good stability across multiple runs, although it was unable to reach the global optimum in all simulations.

Page 46 of 47 | Total Record : 464

 43   44   45   46   47 

Filter by Year

2017 2026


Reset
Filter By Issues
All Issue Vol. 10 No. 1 (2026): Issues July 2026 Vol. 9 No. 2 (2026): Issues January 2026 Vol. 9 No. 1 (2025): Issues July 2025 Vol. 8 No. 3Spc (2025): Special Issues 2025: Innovations in Predictive Analytics and Sentiment Analy Vol. 8 No. 2 (2025): Issues January 2025 Vol. 7 No. 2 (2024): Vol. 7 No. 2 (2024): Issues January 2024 Vol. 8 No. 1 (2024): Issues July 2024 Vol. 7 No. 1 (2023): Issues July 2023 Vol. 6 No. 2 (2023): Issues January 2023 Vol 6, No 1 (2022): Issues July 2022 Vol 5, No 2 (2022): Issues January 2022 Vol 5, No 1 (2021): EDISI JULY 2021 Vol 4, No 2 (2021): EDISI JANUARY 2021 Vol. 4 No. 1 (2020): ---> EDISI JULI Vol 4, No 1 (2020): ---> EDISI JULI Vol 3, No 2 (2020): EDISI JANUARI Vol 3, No 1 (2019): EDISI JULI Vol 2, No 2 (2019): EDISI JANUARI Vol 2, No 2 (2019): EDISI JANUARI Vol 2, No 1 (2018): EDISI JULI Vol 2, No 1 (2018): EDISI JULI Vol 1, No 2 (2018): Edisi Januari Vol 1, No 2 (2018): Edisi Januari Vol 1, No 1 (2017): Edisi Juli Vol 1, No 1 (2017): Edisi Juli More Issue