Claim Missing Document
Check
Articles

Found 33 Documents
Search

Machine Learning-based Chatbot Model for Healthcare Service: A Bibliometric Analysis Nia Ekawati; Imam Riadi; Herman Yuliansyah
Aviation Electronics, Information Technology, Telecommunications, Electricals, and Controls (AVITEC) Vol 7, No 3 (2025): November (Special Issue)
Publisher : Institut Teknologi Dirgantara Adisutjipto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28989/avitec.v7i3.3050

Abstract

While machine learning-based chatbots hold significant potential in healthcare services, a comprehensive synthesis regarding their roles, user demographics, benefits, and limitations remains unavailable, hindering in-depth understanding and future development. This study aims to conduct a bibliometric analysis to identify implementation trends and the research landscape of ML-based chatbot models in healthcare, simultaneously highlighting relevant existing gaps. Analysis of Scopus data using VOSviewer and “Publish or Perish” reveals “machine learning”, “chatbot” and “healthcare” as dominant keywords, indicating intensive research focus areas with stable publication growth. The United States emerges as a central hub for international research collaboration, particularly in AI for malnutrition; however, several outlier countries require further integration. Deep learning algorithms are identified as a crucial methodological trend for future directions. Chatbots possess the potential to revolutionize healthcare by enhancing accessibility and efficiency. Nevertheless, effective implementation necessitates careful consideration of ethical aspects, privacy, and data quality. The identified research gaps underscore the urgency for a holistic synthesis to guide responsible and effective chatbot innovation.
Development of Integrated Project-based (PjBL-T) model to improve work readiness of vocational high school students Bambang Sudarsono; Fatwa Tentama; Surahma Asti Mulasari; Tri Wahyuni Sukesi; Sulistyawati Sulistyawati; Fanani Arief Ghozali; Herman Yuliansyah; Lu'lu' Nafiati; Herminarto Sofyan
Jurnal Pendidikan Vokasi Vol. 12 No. 3 (2022)
Publisher : ADGVI & Graduate School of Universitas Negeri Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21831/jpv.v12i3.53158

Abstract

The unemployed from vocational high schools (SMK) is still high. Several factors, including the low job readiness of vocational students and the small availability of jobs, cause the high unemployment rate for vocational schools. This study aims to develop a learning model of Integrated Project Based Learning (PjBL-T) and test the effectiveness of the PjBL-T model in preparing vocational students' job readiness. The design of this study adopted the Richey and Klein model stages. This research was carried out in three stages: model development, internal validation, and external validation. This study used research subjects consisting of 10 vocational teachers, ten industrial practitioners, and 54 Automotive Engineering SMK Muhammadiyah 2 Tempel students. The research objects used were SMK Muhammadiyah 2 Tempel, Jogjakarta Automotive Center (OJC) Auto Service, Barokah Auto Service, Astra Daihatsu Armada, and Gadjah Mada Auto Service. The data collection techniques used were focus group discussions (FGD), questionnaires, and practice assessment sheets. Data were analyzed descriptively. The PjBL-T model is feasible and can be applied according to the learning objectives. The effectiveness of increasing student work readiness tested limited and expanded and improved very well with a score of 3.27.
Spatio-Temporal Graph-Based Hotspot Analysis of Earthquake Events Using Spatial Autocorrelation and Community Detection in Indonesia Ika Arfiani; Herman Yuliansyah; Nur Rochmah Dyah Puji Astuti; Arfiani Nur Khusna
Journal of Information System and Informatics Vol 8 No 3 (2026): June
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v8i3.1641

Abstract

Analysis of clustered seismic regions is important for understanding seismic activity patterns in tectonic regions such as Indonesia. However, conventional spatial statistical approaches generally analyze earthquake events independently and fail to capture complex spatio-temporal relationships. This study proposes a graph-based spatio-temporal hotspot analysis approach integrating spatial autocorrelation and community detection to identify regional seismic interaction patterns. The dataset used consists of 3,000 earthquake events from 2008–2025. Spatial autocorrelation was analyzed using Moran’s I, while earthquake relationships were modeled using a spatio-temporal graph with spatial and temporal thresholds of ≤400 km and ≤60 days. The results showed significant positive spatial autocorrelation with Moran’s I = 0.3367 (p = 0.001). The resulting graph consisted of 3,000 nodes and 22,896 edges, revealing substantial regional-scale connectivity and 14 major clusters with a modularity score of 0.7405, indicating a strong community structure. Degree centrality analysis identified highly connected nodes with a maximum degree of 77. These findings indicate that integrating spatial autocorrelation and graph analysis provides a more comprehensive representation of seismic interaction patterns and may support future seismic risk assessment in tectonically active regions.
PELATIHAN PEMANFAATAN GENERATIVE ARTIFICIAL INTELLIGENCE UNTUK LITERASI DIGITAL MAHASISWA STIDKI AL-AZIZ BATAM Pastima Simanjuntak; Rika Harman; Yohanni Syahra; Herman Yuliansyah; Imam Riadi
PUAN INDONESIA Vol. 8 No. 1 (2026): Jurnal PUAN Indonesia Vol. 8 No. 1 Juli 2026
Publisher : ASOSIASI IDEBAHASA KEPRI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37296/jpi.v8i1.516

Abstract

The development of Generative Artificial Intelligence (AI) presents significant opportunities to support digital literacy and creativity in higher education. However, utilization of this technology among students still faces challenges, including limited understanding, technical skills, and ethical awareness. This community service activity was conducted on January 31, 2026, at STIDKI Al-Aziz Batam, involving 18 student participants from two study programs. The activity employed an educative-participatory approach comprising material presentations, interactive discussions, and hands-on practice sessions. Post-training evaluation was conducted using a Likert-scale questionnaire (scale 1-5) covering three dimensions: conceptual understanding, perceived benefits, and participant satisfaction. Results showed a mean score of 4.21 out of 5 (84.2%) for conceptual understanding of Generative AI, a mean perceived benefit score of 4.35 (87.0%), and an overall satisfaction score of 4.40 (88.0%). Furthermore, 88.9% of participants (16 of 18) met the minimum understanding threshold (score >= 4). These findings demonstrate that structured, practice based training significantly enhances students digital literacy competence and ethical awareness of AI use. The activity provides a tangible contribution to strengthening students digital capacities and is relevant for sustainable development.
Perbandingan Kinerja MobileNetV2 dan VGG16 dalam Klasifikasi Penyakit pada Citra Daun Tanaman Cabai Itsnaini Irvina Khoirunnisa; Abdul Fadlil; Herman Yuliansyah
JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 11 No. 1 (2026): January 2026
Publisher : UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/jiska.5075

Abstract

Chili peppers play a crucial role in the Indonesian economy, serving as a significant source of income for many farmers. Price fluctuations influenced by weather conditions make this crop vulnerable to diseases that can impact productivity. However, leaves are key indicators of plant health, revealing early disease symptoms before they spread. This research focuses on detecting diseases in chili plants using neural network architectures via transfer learning, specifically MobileNetV2 and VGG16, to classify chili leaf images. The study aims to identify three disease classes: begomovirus, leaf spots, and healthy leaves. The dataset comprises 3,150 leaf images, split into 70% for training and 30% for testing. Results show that MobileNetV2 achieved an accuracy of 99.47% and VGG16 98.62%, with evaluation using a confusion matrix indicating good performance in disease identification, where MobileNetV2 offers better computational efficiency. Thus, transfer learning can effectively identify leaf diseases in chili plants.
An Indonesian Mental Health Chatbot Model Based on A Sequence to Sequence LSTM Nia Ekawati; Nia Ekawati; Imam Riadi; Herman Yuliansyah
JUITA: Jurnal Informatika JUITA Vol. 14 Issue 2, July 2026
Publisher : Department of Informatics Engineering, Universitas Muhammadiyah Purwokerto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30595/juita.v14i2.30040

Abstract

This research proposes an Indonesian-language mental health chatbot model based on the LSTM Sequence-to-Sequence (Seq2Seq) architecture as an adaptive initial support solution. Unlike static classification models, this generative approach aims to capture emotional dependencies and conversational context through context vectors. The research methodology utilizes the public PSYCHIKA dataset, which includes 5,667 conversation pairs a significant volume for a low-resource language. Evaluation was conducted by comparing 80:20 and 70:30 data split schemes. Experimental results showed the best performance with the 80:20 split, achieving a BLEU-1 score of 0.137, compared to the 70:30 split, which only reached 0.043. The model achieved stable convergence at 15–16 epochs via an early-stopping mechanism without any signs of overfitting. Although training stability was maintained, the low BLEU score confirms that the use of a pure Seq2Seq LSTM without an attention mechanism is not yet sufficient to generate highly fluent responses. These findings provide a reproducible technical baseline for the development of mental health dialogue systems in Indonesia, while also emphasizing the urgency of more advanced architectures to improve the quality of empathy in the future.
Model Klasifikasi Emosi Berbasis Teks dengan Algoritma Decision Tree dan Support Vector Machine Habib Aulia Raihan; Herman Yuliansyah; Murinto
Jurnal Informatika dan Rekayasa Perangkat Lunak Vol. 7 No. 2 (2025): September
Publisher : Universitas Wahid Hasyim

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

Abstract

Text-based communication has become a key means of interaction across various sectors. Previous studies have applied supervised learning algorithms to emotion classification in text. These studies used different datasets, but this diversity also introduced a risk of overfitting in text-based emotion classification models. Consequently, the use of cross-validation and hyperparameter optimization is required to ensure the model’s generalization ability. The aim of this research is to compare the performance of two supervised learning algorithms—Decision Tree (DT) and Support Vector Machine (SVM)—for emotion classification on an English-language text dataset of 16,000 labeled entries (anger, fear, joy, love, sadness, surprise) sourced from Kaggle. The dataset undergoes cleaning, tokenization, stopword removal, and lemmatization, after which features are extracted using TF-IDF. Both algorithms are evaluated with K-Fold and Stratified K-Fold cross-validation, then used to compute metrics of accuracy, precision, recall, and F1-score. Classification results show that the hyperparameter-tuned DT achieved an average accuracy of 88%, while the hyperparameter-tuned SVM achieved 89%. Meanwhile, Stratified K-Fold cross-validation yielded an accuracy variance of just 0.02% for DT and 0.15% for SVM. Therefore, it can be concluded that Stratified K-Fold performs better than standard K-Fold on imbalanced datasets, and that hyperparameter-tuned SVM outperforms hyperparameter-tuned DT.
Komparasi Kinerja Algoritma Machine Learning Untuk Analisis Sentimen Multi-Kelas Pada Data Teks Digital Terhadap Kebijakan Pemerintah Sestri Noviarizki; Sunardi Sunardi; Herman Yuliansyah
INTEGER: Journal of Information Technology Vol 11, No 2 (2026): September
Publisher : Fakultas Teknologi Informasi Institut Teknologi Adhi Tama Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31284/j.integer.2026.v11i2.9346

Abstract

The development of social media and digital platforms has increased the number of public opinions on various government policies, including the Free Nutritional Meals (MBG) program. Digital text data generated by the public contains important information regarding public perceptions and sentiments, however  the characteristics of unstructured data and large data volumes pose challenges in the analysis process. In addition, the integrity of class sentiment can also affect the performance of model classification. This study aims to compare the performance of several Machine Learning algorithms, namely Naive Bayes, Support Vector Machine (SVM), Random Forest, Decision Tree, and K-Nearest Neighbors (KNN) in multi-class sentiment analysis classification on digital text data related to the MBG program. The dataset used is 4,969 text data from Kaggle in the form of public comments on social media related to government policies. The research method is through data collection, text preprocessing, TF-IDF feature weighting, model training, and evaluation using a confusion matrix, accuracy, precision, recall, and F1-score. The preprocessing stages include case folding, tokenizing, stopword removal, and stemming to clean and normalize the text data. Sentiment classification was performed into three categories: positive, negative, and neutral. The results showed that the Random Forest algorithm performed best with an accuracy of 0.81, compared to SVM (0.77), Decision Tree (0.75), Naive Bayes (0.71), and K-Nearest Neighbors (KNN) (0.52). Furthermore, Random Forest also produced higher precision, recall, and F1-score values for most sentiment classes, making it more effective in recognizing sentiment patterns in digital text data on government policies related to the MBG program.
Subject Area Classification of Journal Articles Based on Metadata Using Bag of Words and Naïve Bayes Ainunna’imah; Herman Yuliansyah; Imam Riadi
Engineering Science Letter Vol. 5 No. 02 (2026): Engineering Science Letter
Publisher : The Indonesian Institute of Science and Technology Research

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56741/IISTR.esl.002041

Abstract

The rapid growth of scientific publications poses challenges in grouping journal articles based on subject area, especially when using metadata such as titles, abstracts, and keywords. However, differences in feature representation and classification algorithms often result in varying performance, requiring comparative studies to determine the optimal model combination. This study compares four combinations of subject area classification models, namely TF-IDF + Naïve Bayes, TF-IDF + Support Vector Machine, Bag-of-Words + Support Vector Machine, and Bag-of-Words + Naïve Bayes. The research process included text preprocessing, feature extraction, and testing using an 80% training and 20% testing data split scheme in five scenarios. The evaluation was performed using confusion matrices, accuracy, precision, recall, and F1-score. The experimental results showed variations in performance between models, with an average F1-score of 0.8103 for TF-IDF + Naïve Bayes, 0.8494 for TF-IDF + Support Vector Machine, 0.8297 for Bag-of-Words + Support Vector Machine, and 0.8335 for Bag-of-Words + Naïve Bayes as the best performance. These findings indicate that a word frequency-based approach combined with Naïve Bayes is effective for classifying journal article subject areas based on metadata, although challenges remain in subject areas with semantic proximity.
Performance Comparison of SMOTE-Based Ran-dom Forest and Decision Tree Models for Chronic Kidney Disease Classification Rahmat Dwi Cahyono; Herman Yuliansyah; Sunardi
Engineering Science Letter Vol. 5 No. 03 (2026): Engineering Science Letter - Articles in Press
Publisher : The Indonesian Institute of Science and Technology Research

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56741/IISTR.esl.002553

Abstract

Chronic Kidney Disease (CKD) is a chronic condition characterized by a gradual decline in kidney function and is often detected at an advanced stage. Previous studies have employed machine learning methods-such as Random Forest and Decision Trees-for CKD classification; however, there is no consensus regarding which algorithm is most optimal for CKD datasets. The large number of clinical variables in CKD data increases computational complexity, necessitating high-performance machine learning methods to improve classification accuracy and support early disease detection. This study aims to develop a CKD classification model and compare the performance of Decision Tree and Random Forest algorithms. The research utilized a health dataset obtained from Mendeley Data. The methodology involved data acquisition, exploratory data analysis, data preprocessing (including addressing class imbalance via the Synthetic Minority Over-sampling Technique, or SMOTE), and model development using Random Forest and Decision Trees, with evaluation performed via 10-fold stratified cross-validation. Model performance was assessed using confusion matrices and classification reports, covering metrics such as accuracy, precision, recall, and F1-score. Experimental results indicate that Random Forest yielded the best classification performance, achieving an accuracy of 0.9437. It is concluded that comparing Random Forest and Decision Tree methods-combined with SMOTE-is an effective approach for identifying the optimal algorithm to enhance accuracy and support decision-making systems for the early detection of CKD.