cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Hairani
Contact Email
matrik@universitasbumigora.ac.id
Phone
+6285933083240
Journal Mail Official
matrik@universitasbumigora.ac.id
Editorial Address
Jl. Ismail Marzuki-Cilinaya-Cakranegara-Mataram 83127
Location
Kota mataram,
Nusa tenggara barat
INDONESIA
MATRIK : Jurnal Manajemen, Teknik Informatika, dan Rekayasa Komputer
Published by Universitas Bumigora
ISSN : 18584144     EISSN : 24769843     DOI : 10.30812/matrik
Core Subject : Science,
Computer Science & IT
MATRIK adalah salah satu Jurnal Ilmiah yang terdapat di Universitas Bumigora Mataram (eks STMIK Bumigora Mataram) yang dikelola dibawah Lembaga Penelitian dan Pengabadian kepada Masyarakat (LPPM). Jurnal ini bertujuan untuk memberikan wadah atau sarana publikasi bagi para dosen, peneliti dan praktisi baik di lingkungan internal maupun eksternal Universitas Bumigora Mataram. Jurnal MATRIK terbit 2 (dua) kali dalam 1 tahun pada periode Genap (Mei) dan Ganjil (Nopember).
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 454 Documents
 42   43   44   45   46 
Topic Modeling Analysis of Indonesia Food-Security News: Methods,Interpretations, and Trend Insights Afiyati Afiyati; Imbuh Rochmad; Setiyo Budiyanto; Bambang Jokonowo; Hadi Santoso; Kelik Budiana
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 25 No. 2 (2026)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v25i2.5784

Abstract

The critical problem for food-security stakeholders in Indonesia is the lack of scalable, quantitative methods to systematically distill dominant themes and evolving trends from vast volumes of news media, which severely hinders timely policy monitoring and responsive intervention. This study aimed to develop and validate a reproducible topic modeling pipeline specifically designed to uncover the latent thematic structure and quantify the temporal dynamics within Indonesian food-security news discourse. The research method is a comprehensive natural language processing pipeline applied to a curated corpus of 770 news documents spanning 2012 to 2025. The process involved languageadaptive preprocessing of Indonesian text, n-gram (1-2) vectorization to capture nuanced phrases, and training multiple Latent Dirichlet Allocation (LDA) models. The optimal model, with K=10 topics,was rigorously selected through a perplexity-based grid search across a range of potential topic numbers. The resulting topics were then qualitatively interpreted and manually labeled into policy-relevant themes by domain experts. Subsequently, we computed monthly topic intensity series to conduct a longitudinal analysis. The results of this research are that the pipeline successfully generated semantically coherent topics that aligned perfectly with core policy pillars, including availability, access, and utilization. Furthermore, the analysis revealed significant temporal shifts, sustained intensification of price and inflation-related discussions throughout the 2022-2024 period. This study conclusively demonstrates that unsupervised topic modeling can effectively transform unstructured news streams into actionable, quantifiable intelligence, thereby significantly enhancing situational awareness and supporting evidence-based decision-making for food security stakeholders.
Identification of the Sub-motifs of Batik Kawung Using Deep Learning Budi Sunarko; Subiyanto Subiyanto; Hari Wibayanto Wibawanto; Alfanza Rizky Zakaria Zakaria; Alifian Alifian; Naufal Muhammad Muhammad; Yudha Andriano Rismawan Rismawan
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 25 No. 2 (2026)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v25i2.5818

Abstract

Batik is one of Indonesia’s cultural heritages, with motifs that are both diverse and intricate. The Kawung motif, characterized by repetitive circular patterns, is divided into sub-motifs such as Kawung Bribil, Kawung Sen, and Kawung Picis. Automatic classification of these sub-motifs is important for digital preservation but remains difficult due to subtle inter-class similarities. The aim of this research is to analyze the performance of VGG, ResNet, and DenseNet and determine the most effective CNN architecture in classifying the sub-motifs of Batik Kawung. The research method is a convolutional neural network-based image classification approach using a dataset of 300 Kawung Batik images evenly distributed across three classes. Preprocessing steps included grayscale conversion, resizing to 256 × 256 pixels, Canny edge detection, and normalization to the range [0,1]. The dataset was randomly split into 210 training, 60 validation, and 30 testing images. The results of this research are that VGG achieved the highest training accuracy of 97%, but only 67% on the testing set, indicating a tendency to overfit. In contrast, DenseNet achieved the best generalization performance with a testing accuracy of 80%, surpassing both VGG and ResNet. At the class level, DenseNet161 demonstrated consistent performance across all Kawung sub-motifs, with precision ranging from 67% to 91% and F1-scores between 71% and 95%. These results suggest that DenseNet161 not only performed effectively during training but also generalized well to unseen data, establishing it as the most robust architecture for sub-motif Batik Kawung classification. The results underscore the effectiveness of CNNs, particularly DenseNet, in classifying subtle batik sub-motifs. This research contributes to develope a reliable automated system for identifying Kawung batik, leveraging modern technology to support the preservation of Indonesia’s cultural heritage.
Comparative Analysis of Indonesian Pre-trained BERT Models for the Extractive Question Answering Task on an Indonesian-Translated SQuAD Dataset Fattah Al Ilmi Suhendra; Astie Darmayantie; Adang Suhendra Suhendra; Pa Pa Min
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 25 No. 2 (2026)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v25i2.5847

Abstract

Transformer-based architectures have significantly advanced Natural Language Processing (NLP), with Bidirectional Encoder Representations from Transformers (BERT) serving as a strong baseline for extractive Question Answering (QA). This study aims to evaluate the performance of Indonesian BERT models on extractive QA tasks and to identify the most effective model for low-resource language settings. This research employed a comparative experimental method using two Indonesian BERT variants: indobert-base- ncased (IndoLEM) and indobert-base-p1 (IndoNLU/IndoBenchmark). Both models were fine-tuned on an Indonesian version of SQuAD 2.0, automatically translated via the Google Translate API. Answer-span alignment errors caused by translation were corrected using fuzzy string matching. Evaluation was conducted under identical hyperparameter settings and training schemes, using Exact Match (EM) and F1-score as performance metrics. The results indicate that IndoLEM achieved superior performance, with better loss convergence and a higher F1-score (71.58) than IndoNLU (63.59), and the difference was statistically significant (p < 0.001). In conclusion, IndoLEM is a more effective baseline model for Indonesian extractive QA systems. The findings also demonstrate that the composition and scale of pre-trained corpora substantially influence model performance in low-resource language contexts and highlight the importance of transfer learning for advancing NLP in underrepresented languages.
Performance Comparison of LSTM, XGBoost, and Residual-Correction Hybrid LSTM–XGBoost Models for Bitcoin Price Forecasting Ihsan Maulana Anwas; Feri Fahrianto; Imam Marzuki Shofi; Ajif Yunizar Pratama
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 25 No. 2 (2026)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v25i2.5983

Abstract

The objective of this study is to systematically compare the predictive performance of Long Short- Term Memory (LSTM), Extreme Gradient Boosting (XGBoost), and a Hybrid LSTM–XGBoost model for next-day Bitcoin (BTC–USD) closing-price forecasting. The research method employs a quantitative time-series modeling approach using a decade-long daily Bitcoin price dataset. A strictly chronological train–test split and a one-step-ahead forecasting scheme are applied to prevent lookahead bias and ensure experimental validity. Model performance is evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), symmetric Mean Absolute Percentage Error (sMAPE), Mean Absolute Percentage Error (MAPE), and the coefficient of determination R2 on the original price scale. The results demonstrate that the Hybrid LSTM–XGBoost model consistently outperforms the standalone LSTM and XGBoost models across all evaluation metrics, indicating superior predictive accuracy and robustness under high market volatility. The contribution of this study lies in providing a controlled, uniform, and methodologically rigorous head-to-head comparison of deep learning, machine learning, and hybrid architectures for Bitcoin price forecasting, thereby enriching the empirical literature and offering a reliable foundation for the development of adaptive decision-support systemsin volatile cryptocurrency investment environments.
Optimizing Content Recommendations Using a Hybrid Filtering Algorithm to Enhance User Relevance and Engagement Lusiana Efrizoni; Junadhi; Agustin
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 25 No. 2 (2026)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v25i2.5988

Abstract

Recommender systems play an important role in helping users discover relevant content in environments characterized by information overload. However, existing approaches often struggle to balance recommendation relevance and user engagement. Collaborative filtering is constrained by data sparsity and the cold-start problem, whereas content-based methods that rely on textual features may not fully capture dynamic user preferences. This study aims to develop a hybrid deep learning-based recommendation model that improves both recommendation relevance and user engagement. The proposed method integrates collaborative filtering via Neural Matrix Factorization (NeuMF) with content-based filtering via a Long Short-Term Memory (LSTM) text encoder, employing an early-fusion strategy. An experimental research method was applied using synthetic user–item interaction data. Model performance was evaluated using ranking metrics (Precision@10, Recall@10, and NDCG@10) and engagement metrics (Click-Through Rate and Average Completion Ratio). The results show that the hybrid model outperforms the baseline models. It achieves Precision@10 of 0.143, Recall@10 of 0.112, and NDCG@10 of 0.139, which exceed those of both the NeuMF-only and LSTM-only models. In terms of engagement, the hybrid model also records the best performance with a CTR of 0.0017 and an ACR of 0.0090. These findings indicate that integrating user–item interaction patterns with semantic content representations can significantly enhance recommendation quality and user engagement, providing a more effective solution for content-rich digital platforms.
Multi-Criteria Hypervisor Selection Using Analytic Hierarchy Process with Ex-Post Evaluation Ronaldo Agung Nugroho; Dana Indra Sensuse; Sofian Lusa
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 25 No. 2 (2026)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v15i2.6062

Abstract

reassessment of virtualization platform selection in banking Information Technology environments. The objective of this study is to develop a structured, scalable decision-making model to determine the most appropriate hypervisor platform based on technical and non-technical criteria. The research method used is the Analytic Hierarchy Process, developed from qualitative coding of expert interview results and validated through pairwise comparisons by internal infrastructure specialists. The analysis includes consistency measurements, sensitivity analyses, and an ex-post evaluation by comparing analytical ranking results with actual organizational decisions. The results show that technical criteria dominate the decision process, accounting for 64.10% of the total decision weight. At the alternative level, the final priority weights are 45.40–45.44% for Alternative 1, 38.00% for Alternative 2, and 16.55–16.60% for Alternative 3, with Alternative 1 identified as the most optimal choice. Notably, the proposed model achieves a 100% alignment between the analytical ranking and the actual organizational decision, representing a substantial improvement over prior studies, which were largely confined to ex-ante evaluations and lacked empirical validation of decision outcomes. The conclusion of this study confirms that integrating ex-post evaluation into a multi-criteria decision analysis approach enhances the validity of the results and demonstrates a strong fit between the analytical model and real-world decision-making in the context of banking information technology infrastructure.
Lightweight and Interpretable Coin Recognition and Counting UsingGeometric Detection and Fuzzy Score-Based Classification Ni Gusti Ayu Dasriani; Bambang Krismono Triwijoyo; I Gede Yoga Sudarma Yasa; Dadang Priyanto; Cong Dai Nguyen
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 25 No. 2 (2026)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v25i2.6067

Abstract

Deep learning-based coin recognition approaches typically require large, annotated datasets and substantial computational resources, yet offer limited interpretability. Such characteristics limit their applicability in lightweight, resource-constrained vision systems. Therefore, this study aims to develop and systematically evaluate a lightweight, interpretable coin recognition and counting method based on geometric detection and fuzzy-score-based classification. The main contribution of this work lies in integrating the Hough Circle Transform, contour-based circularity validation, and a weighted fuzzy score mechanism that aggregates diameter, circularity, and HSV color features without relying on data-driven model training. The proposed approach prioritizes computational efficiency and decision transparency, while maintaining robustness under varying lighting and object configurations. An experimental evaluation was performed on 40 test images containing 362 coins under both bright and dim lighting conditions, with aligned, scattered, and overlapping arrangements. The system achieved a detection rate of 87% and an object-level classification accuracy of 79%. Although image-level accuracy reached 50% under strict evaluation criteria, detailed error analysis indicates that performance degradation is primarily associated with segmentation limitations in overlapping configurations rather than instability in the fuzzy scoring mechanism. These findings demonstrate that a calibrated geometric and fuzzy-based approach can provide a transparent and computationally efficient alternative for small-scale vision applications without requiring large training datasets.
A Hyperparameter-Adaptive Multilayer Perceptron Framework for Revenue Prediction Based on E-Commerce User Behavior Data Safrizal; lili Tanti; Yan Yang Thanri
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 25 No. 3 (2026)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v25i3.5866

Abstract

Accurate revenue prediction remains a critical challenge for e-commerce platforms due to the highly nonlinear and dynamic nature of user behavior. At the same time, many existing machine learning approaches rely on static model configurations that limit predictive robustness. Although various techniques have been proposed for e-commerce revenue prediction, a systematic, performance-driven approach to adapting Multilayer Perceptron hyperparameters remains underexplored. This study proposes a hyperparameter-adaptive Multilayer Perceptron framework for predicting e-commerce revenue based on user behavior data. Revenue prediction is formulated as a binary classification problem, where outcomes are categorized into conversion and non-conversion events. The dataset comprises 12,330 e-commerce user sessions with behavioral and contextual features, including page interactions, session duration, bounce rate, and visitor characteristics. The proposed framework employs iterative hyperparameter adaptation by evaluating multiple MLP configurations with variations in network depth, activation functions, optimization algorithms, and regularization levels. Model performance is assessed using accuracy, precision, recall, F1-score, and Area Under the Curve. Experimental results indicate that the configuration with the Adam optimizer, ReLU activation, and moderate regularizationachieves the best performance, yielding 88.93% accuracy and an AUC of 0.91. These findings confirm that hyperparameter-adaptive selection significantly enhances prediction performance compared to static model settings. The proposed framework provides a systematic approach to improving revenue prediction accuracy and offers valuable insights for data-driven decision-making and strategic planning in e-commerce environments.
A Stacking Ensemble Learning Framework for Analyzing SkillsMismatch in IT Graduate Employability Rahmaddeni; Junadhi; Sukri Adrianto; Suandi Daulay; Syarfi Azis; Deshinta Arrova Dewi
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 25 No. 3 (2026)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v25i3.5991

Abstract

The increasing gap between academic outcomes and labor market demands has led to a significant skills mismatch among Information Technology graduates. The purpose of this research is to develop a stacking ensemble-based decision-support framework for analyzing and predicting employability outcomes in a multidimensional skill context. The method used is a stacking ensemble learning approach, in which multiple base learners are combined and optimized with XGBoost as the meta-learner. The study uses a synthetic dataset of 2,000 records with 31 variables designed to represent realistic employability factors, including academic performance, technical skills, soft skills, certifications, and career preferences. To enhance interpretability, SHAP (Shapley Additive exPlanations) is employed to identify the contribution of each feature to the prediction outcomes. The result of this study isthat the proposed stacking framework achieves superior performance compared to individual models, demonstrating improved predictive accuracy and robustness. The analysis further reveals that GPA, technical competencies, soft skills, and professional certifications strongly influence employability. In conclusion, the proposed framework not only improves prediction performance but also provides interpretable insights that support decision-making. These findings offer practical implications for higher education institutions and policymakers in designing curriculum strategies and targeted training programs to reduce skills mismatch and enhance IT graduate employability.
Assessing the Effectiveness of Statistical and Temporal Imputation Methods for Bi-LSTM-Based Forecasting on Environmental and Climate Time Series Data Adelia Desyana Eka Putri; Aji Prasetya Wibawa; Adelia Khansa Ristiaputri; Adhelia Wida Khaidir; Dhia Rafifah Thifal; Agung Bella Putra Utama
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 25 No. 3 (2026)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v25i3.6026

Abstract

Time series data in climatology and environmental research are highly susceptible to missing values that can disrupt temporal structures and degrade forecasting performance. This study evaluates the effectiveness of several imputation methods in improving the predictive performance of a Bidirectional Long Short-Term Memory model across three missing-data mechanisms: Missing Completely at Random, Missing at Random, and Missing Not at Random. The compared methods include mean, median, mode, k-nearest neighbors, multiple imputation by chained equations, and last observation carried forward, with data deletion serving as the baseline. All datasets were normalized using the min–max technique, and model hyperparameters were optimized through Particle Swarm Optimization. Performance was assessed using mean absolute percentage error, root mean square error, and the coefficient of determination. The findings indicate that proper imputation significantly enhances forecasting accuracy compared to deleting incomplete observations. In Dataset 1, the last observation carried forward achieved the best performance with a coefficient of determination of 0.923 and a root mean square error of 3.373. Similarly, Dataset 2 showed optimal results with the same method, producing a coefficient of determination of 0.950 and a root mean square error of 14.458. The most substantial improvement was observed in Dataset 3, where mean imputation reduced the mean absolute percentage error from 3.219 to 0.329 while increasing the coefficient of determination to 0.986. These results highlight the critical role of selecting an imputation strategy in deep learning-based time series forecasting and provide practical guidance for handling incomplete environmental datasets.

Page 44 of 46 | Total Record : 454

 42   43   44   45   46 

Filter by Year

2015 2026


Reset
Filter By Issues
All Issue Vol. 25 No. 3 (2026) Vol. 25 No. 2 (2026) Vol. 25 No. 1 (2025) Vol. 24 No. 3 (2025) Vol. 24 No. 2 (2025) Vol. 24 No. 1 (2024) Vol. 23 No. 3 (2024) Vol. 23 No. 2 (2024) Vol. 23 No. 1 (2023) Vol. 22 No. 3 (2023) Vol. 22 No. 2 (2023) Vol. 22 No. 1 (2022) Vol. 21 No. 3 (2022) Vol. 21 No. 2 (2022) Vol. 21 No. 1 (2021) Vol. 20 No. 2 (2021) Vol. 20 No. 1 (2020) Vol. 19 No. 2 (2020) Vol. 19 No. 1 (2019) Vol. 18 No. 2 (2019) Vol. 18 No. 1 (2018) Vol. 17 No. 2 (2018) Vol. 17 No. 1 (2017) Vol. 16 No. 2 (2017) Vol. 16 No. 1 (2016) Vol. 15 No. 2 (2016) Vol. 15 No. 1 (2015) More Issue