Article Per Year (5 Year)
p-Index From 2021 - 2026
0.751
P-Index
This Author published in this journals
All Journal Building of Informatics, Technology and Science Journal of Social Responsibility Projects by Higher Education Forum Bulletin of Information System Research
Mesran Mesran
Sekolah Tinggi Ilmu Manajemen Sukma, Medan
Aerospace Engineering Agriculture, Biological Sciences & Forestry Automotive Engineering Chemical Engineering, Chemistry & Bioengineering Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Earth & Planetary Sciences Electrical & Electronics Engineering Energy Engineering Industrial & Manufacturing Engineering Materials Science & Nanotechnology Mathematics Social Sciences

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

 1 
Deep Learning-Based Early Detection Optimization for Rice Leaf Diseases to Support Sustainable Local Agriculture Putrama Alkhairi; Agus Perdana Windarto; Mesran Mesran; Roznim Roznim
Building of Informatics, Technology and Science (BITS) Vol 7 No 4 (2026): March 2026
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i4.9338

Abstract

Rice leaf diseases such as Bacterial Blight and Blast are major threats to rice productivity that directly impact food security and the sustainability of local agriculture. This study aims to develop and optimize a deep learning-based early detection system for rice leaf diseases using a Convolutional Neural Network (CNN) architecture, specifically the Inception_v3 model. The research method includes five main stages, namely collecting rice leaf image datasets, data pre-processing (resize, normalization, and augmentation), CNN model design, model training and evaluation, and performance optimization through the application of different optimizer algorithms. Two model variants were tested and compared, namely Inception_v3 Basics with the RMSprop optimizer and Inception_v3 Optimization with the Adam optimizer. Experimental results showed that the Inception_v3 Optimization model provided the best performance, with a Precision value of 0.9672, Recall of 0.8939, F1-score of 0.9291, Balanced Accuracy of 0.9297, Matthews Correlation Coefficient (MCC) of 0.8578, Cohen's Kappa of 0.8573, and AUC ROC of 0.98. These results indicate that the Adam optimizer is able to accelerate convergence and improve model accuracy compared to RMSprop, while producing a more stable and efficient classification system. Thus, this study successfully demonstrated that the optimized Inception_v3 architecture can be used effectively for early detection of rice leaf diseases and has high potential for integration into smart farming systems to support sustainable, technology-based local agricultural practices.
Intervensi Terfase Model IN-ON-IN untuk Mengatasi Gap Implementasi Literasi Koding dan AI Guru SD di Sumatera Utara Akhyar Lubis; Fatma Sari Hutagalung; Riah Ukur Ginting; Mesran Mesran; Juanda Hakim Lubis; Fajrul Malik Aminullah Napitupulu
Journal of Social Responsibility Projects by Higher Education Forum Vol 6 No 3 (2026): March 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/jrespro.v6i3.9752

Abstract

This community service program aimed to address the gap between conceptual knowledge and practical implementation ability of elementary school (SD) teachers in coding and artificial intelligence (AI) instruction in Medan City and Deli Serdang Regency. Initial needs analysis revealed that while many teachers possessed basic familiarity with coding and AI terminology, the majority had not independently designed lesson plans incorporating coding activities nor systematically used coding platforms in classroom settings. Partner schools also faced uneven device availability and internet connectivity, alongside weak post-training support structures that hindered the transfer of learning to classroom practice. The intervention applied the IN-ON-IN model with contextual coaching: In Service Training 1 (conceptual and practical workshops using Scratch 3.0, Code.org, and Teachable Machine), On-the-Job Training (classroom implementation with field coaching and telementoring), and In Service Training 2 (reflection and consolidation). The program involved 175 elementary school teachers from both regions, implemented October 2025 through February 2026. Evaluation used an explanatory sequential mixed-methods design: pretest–posttest, classroom observation rubrics, learning artifact analysis, and in-depth interviews and focus groups. Results showed knowledge score improvement from 81% to 95% with reduced standard deviation (3.11 to 1.09), alongside qualitative findings confirming gains in lesson plan design competency and project-based coding instruction for the majority of participants. Persistent barriers included limited devices, intermittent internet access, and time constraints for lesson planning. Recommendations emphasize low-resource module development, expansion of telementoring, and cross-stakeholder collaboration to strengthen school infrastructure.
Mapping the Research Landscape of Multi-Objective Optimization by Ratio Analysis (MOORA) within Multi-Criteria Decision-Making: A Comprehensive Bibliometric and Science Mapping Analysis from 2012 to 2024 Juni Ismail; Alfry Aristo Jansen Sinlae; Zulfikar Zulfikar; Yanto Saputra; Elsy Rahajeng; Mesran Mesran
Bulletin of Information System Research Vol 3 No 2 (2025): April 2025
Publisher : Graha Mitra Edukasi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62866/bios.v3i2.191

Abstract

The Multi-Objective Optimization by Ratio Analysis (MOORA) method has become an increasingly prominent technique within the multi-criteria decision-making (MCDM) family due to its computational simplicity, mathematical stability, and strong capacity to rank alternatives under conflicting criteria, yet the intellectual structure of this rapidly expanding field remains fragmented and insufficiently mapped. This study aims to systematically chart the global research landscape of MOORA within the MCDM domain and to identify its leading contributors, foundational works, dominant publication outlets, and prevailing thematic structures. A bibliometric research design guided by the PRISMA protocol was adopted, drawing on 275 English-language documents retrieved from the Scopus database for the period 2012 to 2024. The data were analysed using VOSviewer and Scopus analytical tools to examine annual publication trends, subject-area distribution, leading sources, co-citation networks, and keyword co-occurrence patterns. The results reveal a field that has accelerated sharply since 2018 and again after 2020, reaching a peak of seventy-one documents in 2024, with output concentrated in Engineering and Computer Science and disseminated through a heterogeneous ecosystem of mechanical-engineering, cleaner-production, and intelligent-systems outlets. Co-citation analysis confirms a theoretical base anchored in the canonical contributions of Brauers and Zavadskas, while keyword mapping shows MOORA functioning as a central decision-making nucleus closely tied to ratio analysis, optimisation, and surface-roughness applications, with TOPSIS, AHP, and WASPAS emerging as salient companion techniques. The novelty of this study lies in its focused mapping of the MOORA intersection rather than MCDM in general, exposing a loosely integrated thematic structure and a reliance on a narrow citation canon dominated by methodological pioneers. Its principal contribution is a consolidated knowledge map that clarifies the field's foundations and directs future methodological, fuzzy-extension, and interdisciplinary innovation
Random Forest, LSTM, and IndoBERT Comparison for TikTok App Sentiment Analysis Imam Saputra; Mesran Mesran; Ruziana Mohamad Rasli
Building of Informatics, Technology and Science (BITS) Vol 8 No 1 (2026): June 2026
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v8i1.9714

Abstract

The rapid growth of social media platforms like TikTok has generated a massive volume of user reviews on the Google Play Store, serving as a critical indicator of application service quality. However, the unstructured nature of Indonesian social media text and the significant imbalance between sentiment classes pose substantial challenges for automated classification systems. Addressing this class imbalance is highly crucial for application developers, as critical negative and neutral feedback containing essential feature complaints is easily marginalized by the overwhelming majority of positive reviews, leading to biased operational insights. This research conducted a comprehensive comparative study of three distinct computational paradigms: Random Forest, Long Short-Term Memory (LSTM), and IndoBERT, to identify the most effective model for sentiment analysis. A dataset of 5,000 TikTok reviews was meticulously processed using a negation-aware preprocessing pipeline to preserve semantic integrity. To address class imbalance, architecture-specific techniques were deployed, including SMOTE for Random Forest, Class Weighting for LSTM, and Random OverSampling for IndoBERT. The experimental results demonstrate that IndoBERT significantly outperforms other models, achieving the highest global accuracy of 81% and a Macro F1-Score of 0.56. While Random Forest and LSTM yielded lower accuracies of 75% and 71%, respectively, they exhibited stability in predicting the majority class but struggled with the inherent ambiguity of neutral sentiments. The study concludes that IndoBERT’s bidirectional self-attention mechanism provides superior contextual understanding of Indonesian slang and non-formal syntax. This research contributes a robust framework for application developers to monitor public opinion objectively. Furthermore, the findings highlight that despite advanced balancing techniques, the "neutrality bottleneck" remains a challenge, suggesting that future research should explore aspect-based sentiment analysis to enhance classification granularity in the Indonesian NLP domain.
Co-Authors Agus Perdana Windarto Akhyar Lubis Alfry Aristo Jansen Sinlae Elsy Rahajeng Fajrul Malik Aminullah Napitupulu Fatma Sari Hutagalung Imam Saputra Juanda Hakim Lubis Juni Ismail Putrama Alkhairi Riah Ukur Ginting Roznim Roznim Ruziana Mohamad Rasli Yanto Saputra Zulfikar Zulfikar