Article Per Year (5 Year)
p-Index From 2021 - 2026
1.36
P-Index
This Author published in this journals
All Journal IAES International Journal of Artificial Intelligence (IJ-AI) JUITA : Jurnal Informatika Buletin Ilmiah Sarjana Teknik Elektro Journal of Data Insights Journal of Computing Theories and Applications Scientific Journal of Informatics Systematic Literature Review Journal TSURAYYA: Journal of Education, Economy and Religia
Abdul Karim
Hallym University
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Engineering Health Professions Mathematics Nursing Public Health Social Sciences Other

Published : 9 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 9 Documents
Search

 1 
Image-Based Classification of Freshwater Fish Species to Support Feed Recommendation Using Random Forest Hindayati Mustafidah; Suwarsito Suwarsito; Rahmat Setiawan; Abdul Karim
JUITA: Jurnal Informatika JUITA Vol. 13 Issue 2, July 2025
Publisher : Department of Informatics Engineering, Universitas Muhammadiyah Purwokerto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30595/juita.v13i2.27358

Abstract

Accurate identification of freshwater fish species plays a vital role in aquaculture, particularly in determining appropriate feed strategies to optimize fish growth. Visual similarities among species—such as color, shape, and surface texture—often hinder novice farmers from correctly recognizing fish types. This study proposes an image-based classification system using the Random Forest algorithm to identify six freshwater fish species: pomfret (bawal), gourami (gurame), catfish (lele), barb (melem), tilapia (nila), and Java barb (tawes) and provide automated feed recommendations. A total of 120 fish images were used as the dataset, collected from various sources, including online repositories and field documentation. Feature extraction was applied to capture color characteristics (HSV), texture patterns (GLCM), and morphological features (regionprops). The model was trained on 70% of the dataset and tested on the remaining 30%. Evaluation results show that the system achieved a classification accuracy of 83.33%, with a precision of 83.53%, recall of 83.33%, and an F1-score of 82.86%. Notably, catfish, barb, and tilapia classes achieved perfect classification, while pomfret and gourami showed room for improvement due to overlapping visual features. The findings indicate that the integration of Random Forest with multi-domain image features offers an effective, affordable, and practical solution to support the digital transformation of small and medium scale aquaculture systems through intelligent species recognition and feed guidance
Combination of binary particle swarm optimization and random forest for stroke disease prediction Sutikno Sutikno; Rismiyati Rismiyati; Khadijah Khadijah; Abdul Karim
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 15, No 3: June 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v15.i3.pp2290-2299

Abstract

Stroke is a leading cause of death and disability worldwide, making early risk prediction critical for prevention. Machine learning methods such as random forest (RF) have shown strong predictive performance, but accuracy can be further improved through effective feature selection. This research proposes an integrated model that combines binary particle swarm optimization (BPSO) for feature selection with RF for stroke risk classification. Experiments were conducted on two public datasets: the stroke prediction dataset (SPD) and the brain stroke dataset (BSD). Data preprocessing included handling missing values, normalization, and the synthetic minority oversampling technique (SMOTE) to mitigate the minority and majority classes. BPSO was employed to select the most informative features, followed by RF for classification. The BPSO-RF model delivered superior accuracies of 96.13% on the SPD and 96.07% on the BSD, outperforming competing classifiers and feature selection techniques. Important features such as gender, age, work type, residence type, average glucose level, body mass index (BMI), and smoking status were consistently identified as key predictors. These results indicate that integrating swarm intelligence with ensemble learning can effectively improve stroke risk prediction and support clinical decision-making.
An Explainable Multimodal Framework for Chest X-Ray Alert Classification Using Radiology Reports and Images Edy Winarno; Indah Manfaati Nur; Abdul Karim; Saeful Amri; Ismi Elya Wirdati; Prajanto Wahyu Adi
Journal of Computing Theories and Applications Vol. 3 No. 4 (2026): JCTA 3(4) 2026
Publisher : Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62411/jcta.16023

Abstract

Artificial intelligence has the potential to support radiology workflows by assisting in the identification of cases that may require additional clinical attention. However, alert-oriented medical AI systems should provide not only classification outputs but also interpretable evidence that can be reviewed and audited by clinicians. This study develops and evaluates an explainable multimodal framework for binary chest X-ray alert classification using paired radiology reports and chest X-ray images. The text branch employs TF-IDF n-gram features with a class-balanced Logistic Regression classifier, while the image branch fine-tunes a pretrained ResNet18 model. The two branches are integrated through probability-level late fusion using a validation-selected fusion weight. Explainability is implemented in a modality-specific manner: global coefficient analysis is used to identify influential textual cues, while Grad-CAM heatmaps are used to visualize salient image regions. Experiments were conducted on paired samples from the Open-i/IU X-Ray dataset using text-only, image-only, and fusion-based evaluation settings. Additional analyses include case-level complementarity analysis, bootstrap confidence intervals for ROC-AUC, shortcut-feature inspection, and qualitative Grad-CAM auditing. The results indicate that the text modality provides the dominant predictive signal under the current proxy-label setting. Late fusion produced a small descriptive improvement on the test set, increasing accuracy from 0.8533 to 0.8667, F1-score from 0.8817 to 0.8936, and ROC-AUC from 0.8936 to 0.9025 compared with the text-only baseline. However, the observed ROC-AUC improvement was not statistically conclusive based on bootstrap analysis. These findings suggest that the proposed framework is useful as a reproducible and auditable multimodal prototype, while also highlighting important limitations, including proxy-label ambiguity, potential label leakage from radiology reports, limited image-branch contribution, lack of external validation, and the need for stronger explanation and calibration assessment.
INNOVATIVE CURRICULUM DESIGN FOR SUSTAINABLE EDUCATION: BRIDGING LOCAL WISDOM AND GLOBAL CHALLENGES Asmarita; Fitri Yuniarti; Ahmad Afendi; Abdul Karim; Multazom; Suprapno; Moh. Syarif Hidayat
TSURAYYA: Journal Education, Economy and Religia Vol. 1 No. 2 (2026): May: TSURAYYA: Journal Education, Economy and Religia
Publisher : PT. Cadas Insan Madani

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

In the era of globalization and rapid technological advancement, education systems are increasingly challenged to develop curricula that are responsive to both local contexts and global issues. This study explores innovative curriculum design for sustainable education by integrating local wisdom with global challenges. Local wisdom represents valuable cultural knowledge, traditions, and practices that have been developed by communities over generations and contribute to social cohesion, environmental stewardship, and cultural identity. Meanwhile, global challenges such as climate change, digital transformation, social inequality, and sustainable development require learners to possess critical thinking, problem-solving, collaboration, and global citizenship competencies. Through a conceptual and literature-based approach, this paper examines strategies for curriculum innovation that harmonize indigenous knowledge with international educational frameworks, particularly the Sustainable Development Goals (SDGs). The findings indicate that a curriculum grounded in local wisdom while addressing global concerns can foster contextual learning, strengthen cultural preservation, and enhance students’ readiness to participate in an interconnected world. Furthermore, the integration of local and global perspectives promotes sustainable education by encouraging environmental awareness, social responsibility, and lifelong learning. Therefore, innovative curriculum design serves as a strategic framework for creating relevant, inclusive, and future-oriented educational systems that contribute to sustainable development at both local and global levels.
Hate Speech Detection on X Using K-Nearest Neighbor with TF–IDF and Cosine Similarity: Deteksi Ujaran Kebencian pada X Menggunakan K-Nearest Neighbor dengan TF–IDF dan Kesamaan Kosinus Faiq Madani; Arvanida Feizal Permana; Abdul Karim; Riyagung Nuryusufa Tranggono Adi Prasetya; Wendy Sarasjati
Journal of Data Insights Vol 4 No 1 (2026): Journal of Data Insights
Publisher : Department of Sains Data UNIMUS Universitas Muhammadiyah Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26714/jodi.v4i1.1149

Abstract

The rapid growth of social media has increased online interactions but has also accelerated the spread of hate speech content that may negatively impact individuals and communities. X (formerly Twitter), as one of the largest social networking platforms, enables users to share opinions publicly, making automatic hate speech detection increasingly important. This research proposes a hate speech classification approach using the K-Nearest Neighbor (KNN) algorithm combined with Term Frequency–Inverse Document Frequency (TF–IDF) weighting and Cosine Similarity. The dataset consists of 900 social media posts collected through the platform API and manually labeled into hate speech and non-hate speech categories, consisting of 675 training data and 225 testing data. Prior to classification, text preprocessing techniques including tokenization, stopword removal, and stemming were applied to improve text quality. Model evaluation was conducted using 10-fold cross validation to assess classification performance. Experimental results showed that the KNN algorithm with Cosine Similarity distance measurement and K=3 parameter achieved an accuracy of 78.22% in hate speech detection tasks. The findings indicate that KNN combined with TF–IDF and Cosine Similarity provides a reliable approach for social media text classification and can support automated hate speech detection systems.
Advances in Brain-Computer Interfaces for Taste Perception: Current Insights and Future Directions Yuri Pamungkas; Abdul Karim; Gao Yulan; Muhammad Nur Afnan Uda; Uda Hashim
Buletin Ilmiah Sarjana Teknik Elektro Vol. 8 No. 1 (2026): February
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v8i1.14718

Abstract

Human taste perception is a complex multisensory process that integrates chemical, emotional, and cognitive responses within the brain. Traditional methods for evaluating taste rely on subjective reporting, which limits reproducibility and accuracy. Brain-Computer Interface (BCI) technology provides an objective solution by decoding neural activity associated with taste perception using non-invasive techniques such as EEG and fNIRS. The research contribution aims to deliver an extensive overview of the latest advancements in BCI-oriented taste research, emphasizing various applications, methodological frameworks, and potential future pathways that connect the domains of neuroscience and sensory technology. This review examines the use of EEG and fNIRS modalities for signal acquisition, preprocessing, feature extraction, and classification across 36 studies conducted between 2020 and 2025. These works employ both traditional algorithms and deep learning models, including SVM, CNNs, and Transformer-based frameworks, to decode neural signatures of basic tastes and multisensory interactions. Results show that BCIs have successfully identified distinct brain responses for sweet, sour, salty, bitter, and umami stimuli. They have also been applied in multisensory integration, hedonic evaluation, consumer behavior analysis, clinical diagnosis of taste disorders, and affective monitoring. However, challenges remain in signal noise, dataset standardization, and model interpretability. In conclusion, BCIs represent a promising and interdisciplinary approach for objectively studying and enhancing human taste perception through the integration of neuroscience, engineering, and artificial intelligence.
Trends and Gaps in Transformer-Based EEG Modeling: A Review of Recent Developments Yuri Pamungkas; Abdul Karim; Myo Min Aung; Muhammad Nur Afnan Uda; Uda Hashim
Buletin Ilmiah Sarjana Teknik Elektro Vol. 8 No. 2 (2026): April
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v8i2.14933

Abstract

In recent years, Transformer-based deep learning architectures have emerged as a powerful paradigm for modeling EEG signals, offering superior capability in capturing spatial–temporal dependencies compared to traditional convolutional or recurrent networks. However, the diversity of model designs, limited dataset generalization, and lack of standardization have created challenges in evaluating their true potential for real-world applications. This review addresses these issues by systematically examining the evolution, performance, and methodological trends of Transformer-based EEG models published between 2022 and 2024, highlighting both achievements and research gaps. The main contribution of this study is to provide a comprehensive mapping and critical analysis of Transformer architectures applied to EEG classification, feature extraction, and signal decoding tasks. Using the Scopus database, a structured search was conducted following specific inclusion criteria (English, peer-reviewed, open-access journal papers from 2022–2024) and a well-defined query combining EEG and Transformer-related keywords. Data from 63 eligible studies were extracted and categorized according to authorship, dataset, architecture type, EEG application, and evaluation metrics. Results show that hybrid Transformer models dominate recent research, achieving accuracies above 90% in tasks such as motor imagery, emotion recognition, seizure detection, and sleep staging. Pure Transformers like ViT and BERT-like models also demonstrate competitive performance but face scalability and interpretability challenges. In conclusion, Transformer-based EEG modeling is advancing rapidly, yet future efforts must focus on model efficiency, explainability, and benchmark standardization to enable broader clinical and real-world adoption.
Combination Of VADER Sentiment Analysis and SEQ Scale For Evaluating the Usability of The Gojek Application Zakiyah Apriliya Budiarti; Tenia Wahyuningrum; Adnan Purwanto; Muhammad Akbar Setiawan; Singgih Setia Andiko; Abdul Karim
Scientific Journal of Informatics Vol. 13 No. 1: February 2026
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v13i1.35956

Abstract

Purpose: This study aims to analyze user sentiment toward the Gojek mobile application using the Valence Aware Dictionary for Sentiment Reasoning (VADER) method and to evaluate the perceived ease of use of the application using the Single Ease Question (SEQ) instrument. Methods: The research data were obtained by scraping 30,000 user reviews of the Gojek application from the Google Play Store. The reviews were processed through text preprocessing, sentiment classification using the VADER method, and subsequent mapping of sentiment polarity scores to a 1–7 SEQ usability scale. A Spearman rank correlation analysis was conducted to examine the relationship between sentiment scores and derived SEQ values. Result: The results indicate that user sentiment toward the Gojek application is predominantly positive, followed by neutral and negative sentiments. The overall average SEQ score is 4.11, suggesting that the application is generally perceived as fairly easy to use. Furthermore, a strong and statistically significant positive association was found between VADER sentiment scores and SEQ usability scores, indicating that more positive sentiment tends to be associated with higher perceived ease of use. Novelty: This study contributes to the literature by empirically integrating sentiment analysis and usability evaluation using VADER and SEQ within the context of an Indonesian super-app. The findings provide practical insights for digital application developers to identify usability strengths and areas for improvement based on large-scale user feedback.
Groundedness in Government Document Chatbots: A Systematic Literature Review and Metric Oriented Analysis Frendy Rumambi; Didik Dwi Prasetya; Triyanna Widiyaningtyas; Abdul Karim
Systematic Literature Review Journal Vol. 2 No. 1 (2026): January : Systematic Literature Review Journal
Publisher : International Forum of Researchers and Lecturers

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70062/slrj.v2i1.277

Abstract

The application of Large Language Models (LLM) in public services encourages government agencies to adopt Retrieval Augmented Generation (RAG)-based chatbots as interfaces for regulatory knowledge and official documents. Although RAG is designed to increase the supportability of answers to authoritative sources, various studies show that this system is still vulnerable to hallucinations, which have the potential to reduce public trust and pose legal risks. This article presents a Systematic Literature Review (SLR) on the use of RAG in government chatbots with a focus on the definition, mitigation strategies, and evaluation of groundedness. The literature search was conducted in the period 2021–2025 through the SpringerLink, Scopus, and Taylor & Francis databases, resulting in 7,947 articles filtered using the PRISMA framework to obtain 100 articles Q1–Q2. Based on eight research questions, this study maps publication trends, document domains, RAG architecture, retrieval strategies, definitions of groundedness, and evaluation metrics used. The SLR results indicate conceptual fragmentation in the definition and measurement of groundedness, with the dominance of text-similarity-based metrics that are inadequate for regulatory contexts. As a conceptual contribution, this article formulates the Semantic Alignment Score (SAS) as a groundedness metric based on semantic alignment, evidence coverage, and entailment relationships, positioned to support the evaluation and auditing of government document chatbots.
Co-Authors . Suwarsito Adnan Purwanto Ahmad Afendi Amri, Saeful Arvanida Feizal Permana Asmarita Didik Dwi Prasetya Edy Winarno Faiq Madani Fitri Yuniarti Frendy Rumambi Gao Yulan Hindayati Mustafidah Indah Manfaati Nur Ismi Elya Wirdati Khadijah Khadijah Moh. Syarif Hidayat Muhammad Akbar Setiawan Muhammad Nur Afnan Uda Multazom Myo Min Aung Prajanto Wahyu Adi Rahmat Setiawan Rismiyati Rismiyati Riyagung Nuryusufa Tranggono Adi Prasetya Singgih Setia Andiko Suprapno Sutikno Sutikno Tenia Wahyuningrum Triyanna Widiyaningtyas Uda Hashim Wendy Sarasjati Yuri Pamungkas Zakiyah Apriliya Budiarti