cover
Contact Name
Suwanto Sanjaya
Contact Email
suwantosanjaya@uin-suska.ac.id
Phone
-
Journal Mail Official
coreit@uin-suska.ac.id
Editorial Address
-
Location
Kab. kampar,
Riau
INDONESIA
Jurnal CoreIT
ISSN : 2460738X     EISSN : 25993321     DOI : -
Core Subject : Science,
Jurnal CoreIT: Jurnal Hasil Penelitian Ilmu Komputer dan Teknologi Informasi published by Informatics Engineering Department – Universitas Islam Negeri Sultan Syarif Kasim Riau with Registration Number: Print ISSN 2460-738X | Online ISSN 2599-3321. This journal is published 2 (two) times a year (June and December) containing the results of research on Computer Science and Information Technology.
Arjuna Subject : -
Articles 12 Documents
Search results for , issue "Vol 11, No 2 (2025): December 2025" : 12 Documents clear
Classification of Apple Tree Leaf Diseases using Pretrained EfficientNetB0 and XGBoost Qohar, Bagus Al; Dullah, Ahmad Ubai; Darmawan, Aditya Yoga; Unjung, Jumanto
Jurnal CoreIT: Jurnal Hasil Penelitian Ilmu Komputer dan Teknologi Informasi Vol 11, No 2 (2025): December 2025
Publisher : Fakultas Sains dan Teknologi, Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/coreit.v11i2.33174

Abstract

The diseases that affect apple tree leaves seriously compromise agricultural production; therefore, early and accurate diagnosis is quite important for good disease control. Machine learning's recent developments have opened fascinating possibilities for automating the detection process and enhancing methods of precision agriculture. This study aims to create a strong classification model that can accurately and efficiently identify various diseases that affect apple tree leaves. The approach combines the pre-trained EfficientNetB0 architecture for feature extraction with the XGBoost model for classification, utilizing the advantages of both deep learning and gradient-boosting methods. With high performance measures including a macro-average precision of 95.86%, recall of 95.44%, and F1 score of 95.64%, the model achieved a classification accuracy of 95.74%. Furthermore, the average ROC-AUC score of 0.9964 emphasizes how well the model differentiates the five disease categories. This work stands out due to its hybrid approach, which integrates a robust pre-trained convolutional neural network (EfficientNetB0) with the XGBoost model. This significantly improves the accuracy of disease classification. This approach presents a novel pathway for precision agriculture, providing a reliable and effective instrument for the automatic identification of diseases in apple orchards.
Integration of G2M Weighting and MOORA in Accurate Decision Making for Best Alternative Selection Setiawansyah, Setiawansyah; Wang, Junhai; Palupiningsih, Pritasari
Jurnal CoreIT: Jurnal Hasil Penelitian Ilmu Komputer dan Teknologi Informasi Vol 11, No 2 (2025): December 2025
Publisher : Fakultas Sains dan Teknologi, Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/coreit.v11i2.36679

Abstract

The goal of the integration of the G2M Weighting and MOORA methods is to produce the best alternative selection decisions that are more accurate and objective. By combining rational criteria weighting through G2M Weighting and alternative evaluation using MOORA, it is hoped that it can reduce bias and increase transparency in decision-making. In addition, this study compares alternative ratings from the application of the MOORA method and other weighting methods. The results of the evaluation and ranking of scholarship recipients using G2M weighting and MOORA, CF candidates managed to occupy the first position with a final score of 0.2727, showing the best performance among all candidates. In second place, UT candidates obtained a score of 0.2630, followed by DF candidates with a score of 0.2445 and SS candidates with a score of 0.2425. This approach makes it a very useful solution in the selection of the best alternatives in a wide range of multi-criteria decision applications. The results of the Spearman correlation test showed that the G2M weighting method had the highest correlation of 0.9879, which showed a very high similarity with the initial rating. The Entropy Weighting and CRITIC methods also showed a strong correlation, of 0.9515 and 0.9636, respectively, although there was slight variation in the alternate sequence. Meanwhile, the MEREC weighting has the lowest correlation of 0.9273, but still shows a very strong relationship. Overall, these results suggest that the G2M method produces rankings consistent with the initial rankings, with variations indicating sensitivity to criterion weighting.
Benchmarking Various Machine Learning Models to Detect Lung Cancer Afrianty, Iis; Afriyanti, Liza
Jurnal CoreIT: Jurnal Hasil Penelitian Ilmu Komputer dan Teknologi Informasi Vol 11, No 2 (2025): December 2025
Publisher : Fakultas Sains dan Teknologi, Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/coreit.v11i2.38590

Abstract

This study benchmarked and evaluated the performance of various machine learning techniques to detect lung cancer using public datasets. The techniques used include Random Forest, Support Vector Machine, Logistic Regression, Multi-Layer Perceptron, C4.5, Bayesian Network, Reptree, Naive Bayes, and P.A.R.T. Evaluation was carried out using metrics such as Accuracy, F-measure, Precision, TPR, ROC, FPR, PRC, and MCC. The results showed that the Support Vector Machine algorithm performed best on balanced dataset distribution, while Random Forest showed stable performance on unbalanced datasets. This study confirms the importance of selecting appropriate algorithms and data distribution to improve lung cancer detection.
A Support Vector Regression Approach for Predicting the Remaining Useful Life of Turbofan Engines Hardiansyah, Muhammad Vio; Insani (Scopus ID: 57190404820), Fitri; Handayani, Lestari; Jasril, Jasril; Sanjaya, Suwanto
Jurnal CoreIT: Jurnal Hasil Penelitian Ilmu Komputer dan Teknologi Informasi Vol 11, No 2 (2025): December 2025
Publisher : Fakultas Sains dan Teknologi, Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/coreit.v11i2.38532

Abstract

Turbofan engines are crucial components in the aviation and manufacturing industries, where estimating the Remaining Useful Life (RUL) has a significant impact on operational efficiency and safety. This study aims to predict the RUL of turbofan engines using the Support Vector Regression (SVR) method, a machine learning approach that has proven effective in modeling nonlinear relationships between variables. Operational data related to turbofan engines include operational parameters, sensors, and maintenance records. The initial stage of this research involves data analysis based on unit number, time, operational control, and sensor parameters. This process begins with preprocessing to initialize the initial data values, normalize, and select sensors that have stagnant values, as these sensors do not affect the machine learning system. Subsequently, regression calculations are performed to compare predicted values and actual values using the Support Vector Regression method optimized with Grid Search Optimization. In this study, testing was conducted with Parameters C [1, 10, 50, 100] and ε [1, 5, 10, 50], resulting in the best model with an RMSE error of 19.56 and MAE of 14.73.
Combining BERT and Graph-Based Ranking for Extractive Summarization of Indonesian News Articles Trisna, I Nyoman Prayana; Vihikan, Wayan Oger; Azizah, Anis Zahra Nur
Jurnal CoreIT: Jurnal Hasil Penelitian Ilmu Komputer dan Teknologi Informasi Vol 11, No 2 (2025): December 2025
Publisher : Fakultas Sains dan Teknologi, Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/coreit.v11i2.37929

Abstract

Automatic text summarization is an effective solution to manage the vast amount of information in the digital age. This study aims to develop an extractive text summarization system for Indonesian news articles using sentence embeddings generated by IndoBERT and mBERT, combined with TextRank and LexRank algorithms for sentence ranking. The dataset used is Indonesian Text Summarization (IndoSum), which contains thousands of manually summarized articles. The research includes data collection, cleaning, preprocessing, embedding extraction, sentence similarity calculation, and ranking using graph-based methods. Model performance was evaluated using ROUGE and BERTScore. The results show that the combination of IndoBERT and LexRank achieved the highest performance with ROUGE-1 score 0.7018 and BERTscore 0.8696. The model was then implemented into a web prototype using Streamlit to allow users to summarize texts interactively. This study contributes to the advancement of automatic summarization technology for the Indonesian language.
Comparison of Various Deep Learning Techniques to Obtain the Best Technique for Detecting Brain Cancer Yanto, Febi; Budianita, Elvia; Wang, Shir Li
Jurnal CoreIT: Jurnal Hasil Penelitian Ilmu Komputer dan Teknologi Informasi Vol 11, No 2 (2025): December 2025
Publisher : Fakultas Sains dan Teknologi, Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/coreit.v11i2.38599

Abstract

This study aims to address the difficulty of comparing deep learning–based brain cancer detection methods due to differences in datasets and parameter settings, which limits the generalizability of previous findings. The purpose of this research is to evaluate the performance of several convolutional neural network (CNN) architectures using identical datasets and experimental configurations to determine the most effective technique for early brain cancer detection. The study builds a comparative framework using the Keras API on TensorFlow, supported by libraries such as NumPy, Pandas, Matplotlib, and Seaborn. All datasets were split into stratified training, validation, and test sets, and preprocessing included resizing images to 224×224 pixels, converting them to 3-channel RGB, normalizing the inputs, and applying data augmentation. CNN architectures, including VGG16, ResNet50, GoogleNet, and AlexNet, were trained with consistent parameter settings, including epoch count, batch size, learning rate optimization, and training protocols. Performance evaluation using accuracy, precision, recall, and F1-score shows that GoogleNet and ResNet50 achieve the highest results across datasets (average >94%), with GoogleNet slightly outperforming ResNet50. AlexNet performs poorly on the Kaggle dataset but shows potential on the private dataset, while VGG16 demonstrates moderate but less consistent performance. The originality of this study lies in providing a unified evaluation framework that enables fair comparison across CNN models, offering valuable insights for selecting optimal architectures for brain cancer detection.
Comparison Of The Performance Of K-Nearest Neighbors And Naive Bayes Algorithms For Stroke Disease Prediction baskoro, baskoro; Novianto, Roby; Triraharjo, Bambang
Jurnal CoreIT: Jurnal Hasil Penelitian Ilmu Komputer dan Teknologi Informasi Vol 11, No 2 (2025): December 2025
Publisher : Fakultas Sains dan Teknologi, Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/coreit.v11i2.37542

Abstract

Purpose: Stroke is a critical global health issue requiring early and accurate prediction to mitigate severe outcomes. This study aims to compare the performance of the K-Nearest Neighbors (KNN) and Naive Bayes algorithms in predicting stroke disease, addressing the challenge of imbalanced datasets and improving prediction accuracy for better clinical decision-making.Methods/Study design/approach: The research followed the CRISP-DM model, utilizing a dataset of 5,110 patient records with 12 attributes from Kaggle. Data preprocessing included handling missing values and normalization. The KNN and Naive Bayes algorithms were implemented using RapidMiner, with performance evaluated through cross-validation, confusion matrices, and ROC-AUC curves.Result/Findings: The KNN algorithm achieved an accuracy of 94.50%, but exhibited low precision (7.89%) and recall (1.20%) for stroke-positive cases due to dataset imbalance. Naive Bayes yielded an accuracy of 88.83% with an AUC of 0.767, demonstrating better probability modeling but similar challenges in minority class detection. Both algorithms highlighted the impact of data imbalance on predictive performance.Novelty/Originality/Value: This study provides a comparative analysis of KNN and Naive Bayes for stroke prediction, emphasizing the need for data balancing and optimization techniques. The findings underscore the potential of these algorithms in healthcare applications while suggesting future improvements through ensemble methods or alternative algorithms like Random Forest.
Evaluation of the Latent Dirichlet Allocation for Modeling News Topics of Nusantara Capital City Kartika, Luh Gede Surya; Putra, Anggara Putu Dharma; Rinartha, Komang; Megawati, Megawati
Jurnal CoreIT: Jurnal Hasil Penelitian Ilmu Komputer dan Teknologi Informasi Vol 11, No 2 (2025): December 2025
Publisher : Fakultas Sains dan Teknologi, Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/coreit.v11i2.33397

Abstract

Research regarding topic modeling on the coverage of the Nusantara Capital City (IKN) in national mass media remains limited. This study aims to not only model IKN-related topics but also rigorously evaluate the Latent Dirichlet Allocation (LDA) model to ensure its robustness for future implementation. The dataset comprises 1,498 news articles gathered from prominent Indonesian online media, specifically Detik (1,050 articles) and Kompas (448 articles). The methodology involves experimental variations of LDA parameters, including document volume, maximum features, and topic count, utilizing the Scikit-learn library. The results indicate that an increase in data volume and feature dimensions significantly correlates with longer computation times and a higher number of epochs required for convergence. Furthermore, the expansion of variables and data volume resulted in more negative log-likelihood values and increased perplexity, suggesting that model complexity challenges predictive precision. A convergence threshold of $1e^{-2}$ was applied to optimize the training cessation point. While this study establishes a baseline for static topic modeling, future research implies the necessity of Dynamic Topic Modeling (DTM) to capture the temporal evolution of topics, a dimension not addressed by the standard LDA model.
Classification of Herbal Leaves using EfficientNetB0 Alda, A. Nurul Aisya; Indra, Dolly; Umar, Fitriyani
Jurnal CoreIT: Jurnal Hasil Penelitian Ilmu Komputer dan Teknologi Informasi Vol 11, No 2 (2025): December 2025
Publisher : Fakultas Sains dan Teknologi, Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/coreit.v11i2.38017

Abstract

The identification of herbal leaves remains a challenging task due to the high morphological and visual similarity among commonly used species, which often leads to misclassification when performed manually. This study addresses the challenge of identifying herbal leaves, namely Sauropus androgynus, Moringa oleifera, Orthosiphon aristatus, Syzygium polyanthum, and Piper betle, which are often difficult to distinguish due to high morphological and visual similarity.The proposed approach utilizes the EfficientNetB0 Convolutional Neural Network architecture and employs a two-stage fine-tuning strategy, combined with data augmentation, to enhance generalization performance. A total of 500 manually collected leaf images were used for training, resized to 224×224 pixels, and augmented through rotation and flipping. Model optimization was performed using the Adam and SGD optimizers. The trained model was evaluated on 235 previously unseen external images to assess robustness. The experimental results demonstrate that the proposed model achieved an overall classification accuracy of 88.94%, with particularly strong performance on leaf classes exhibiting distinctive morphological features, such as Orthosiphon aristatus, which obtained an F1-score of 0.96. However, the model exhibited limitations in distinguishing visually similar classes, especially between Moringa oleifera and Sauropus androgynus, both of which possess compound leaf structures, and performance degradation was observed under varying illumination conditions and complex backgrounds. The novelty of this study lies in the application of an EfficientNetB0-based fine-tuning strategy for multi-class herbal leaf classification using a limited, manually collected dataset, demonstrating its potential for deployment in mobile or other resource-constrained environments to support fast and reliable herbal plant identification.
Data Mining for Analyzing Consumer Segmentation: Identifying Consumer Preference Patterns Using the Fuzzy C-Means Clustering on Halal Products Iskandar (Scopus ID: 55316114000), Iwan; Nazir, Alwis
Jurnal CoreIT: Jurnal Hasil Penelitian Ilmu Komputer dan Teknologi Informasi Vol 11, No 2 (2025): December 2025
Publisher : Fakultas Sains dan Teknologi, Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/coreit.v11i2.38608

Abstract

Halal products are increasingly popular worldwide, not only in Muslim-majority countries but also in non-Muslim nations.  The global halal market exceeds USD 650 million annually, emphasizing the importance of halal certification, particularly in Indonesia as the world’s largest Muslim-majority country. This research aims to cluster consumers of halal meat products by analyzing factors influencing consumer behavior in purchasing certified halal beef and chicken. The study employs the Fuzzy C-Means (FCM) clustering algorithm on 176 respondents’ questionnaire data consisting of 36 parameters. The experiment was performed using Google Colab with a maximum of 1000 iterations, error tolerance of 0.0001, and fuzziness coefficient m = 2.4. Results show that two optimal clusters were formed, with a Partition Coefficient Index (PCI) value of 0.9993, indicating excellent clustering quality. The first cluster consists primarily of young consumers aged 15–24 with lower spending, while the second cluster includes adults aged 35–54 with higher income. Both groups prioritize halal certification and logo visibility when choosing meat products. The findings provide insights for halal product retailers and policymakers to enhance halal product distribution, certification support, and marketing strategies in Indonesia.

Page 1 of 2 | Total Record : 12