cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Hasniati
Contact Email
puslit@kharisma.ac.id
Phone
+62411-871555
Journal Mail Official
kharismatech@kharisma.ac.id
Editorial Address
Jl. Baji Ateka No. 20 Makassar
Location
Kota makassar,
Sulawesi selatan
INDONESIA
Kharisma Tech
Published by Sekolah Tinggi Manajemen Informatika dan Komputer Kharisma
ISSN : 19072317     EISSN : 28100344     DOI : https://doi.org/10.55645/kharismatech.v17i1
Core Subject : Science, Engineering,
Computer Science & IT Engineering
Jurnal Ilmu Komputer merupakan jurnal yang menampung hasil penelitian di bidang informatika dan sistem informasi, mencakup : - Sistem Informasi - Informatika - Teknologi Informasi - Ilmu Komputer - Software Engineering
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 206 Documents
 17   18   19   20   21 
SISTEM KLASIFIKASI SAMPAH ORGANIK, ANORGANIK, DAN BAHAN BERBAHAYA BERACUN (B3) Bradika Almandin Wisesa; Vivin Mahat Putri; Evvin Faristasari; Sirlus Andreanto Jasman Duli; Rahmat Lionza
KHARISMA Tech Vol 21 No 2 (2026): KHARISMATech Journal
Publisher : STMIK KHARISMA Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55645/kharismatech.v21i2.695

Abstract

Penelitian ini mengembangkan sistem klasifikasi sampah real-time untuk sampah organik, anorganik, dan bahan berbahaya beracun (B3) di Jalan Timah Sungailiat, Bangka Belitung, Indonesia menggunakan YOLOv11 untuk mengatasi pencemaran lingkungan akibat pengelolaan sampah yang tidak tepat. Dataset khusus sebanyak 5.000 gambar yang seimbang pada kondisi siang, senja, dan malam dikumpulkan untuk melatih YOLOv11 sehingga mampu menangani variasi pencahayaan. Metodologi meliputi pra-pemrosesan (ubah ukuran 640×640 piksel, normalisasi, augmentasi data), pelatihan YOLOv11 pada dataset seimbang, serta evaluasi menggunakan metrik mean Average Precision (mAP@0.5), presisi, recall, dan F1-score. Sistem mencapai mAP@0.5 sebesar 70%, presisi 69%, recall 70%, dan F1-score 0,70 dengan kecepatan 43 frame per detik (FPS) serta 102 GFLOPs sehingga dapat dijalankan pada GPU kelas menengah. Meskipun akurasi masih moderat karena variasi pencahayaan dan oklusi, kerangka kerja ini menawarkan solusi hemat biaya dan skalabel untuk pengelolaan sampah kota pintar, mengurangi tenaga sorting manual serta mendukung inisiatif daur ulang. Perbaikan mendatang akan fokus pada peningkatan performa malam hari dan penanganan.
EXPLAINABLE MACHINE LEARNING FOR PHISHING URL DETECTION USING SHAP INTERPRETATION M. Hizbul Wathan; Indra Irawan; Better Swengky; Tri Agusti Farma; Satria Agus Darma
KHARISMA Tech Vol 21 No 2 (2026): KHARISMATech Journal
Publisher : STMIK KHARISMA Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55645/kharismatech.v21i2.696

Abstract

Phishing attacks delivered through malicious URLs represent an increasingly prevalent cyber threat capable of causing significant harm to users. Although machine-learning–based phishing detection has been widely explored, most existing models still operate as black boxes, making their classification decisions difficult to interpret. This study proposes an Explainable Machine Learning framework for phishing URL detection by integrating five algorithms—XGBoost, Random Forest, Gradient Boosting, Decision Tree, and K-Nearest Neighbors—augmented with SHAP (SHapley Additive Explanations) for interpretability. The dataset includes structural URL features such as character length, special symbol counts, number of subdomains, and string entropy. Model performance was evaluated using accuracy, precision, recall, F1-score, and confusion matrix to enable comparative assessment among algorithms. The results show that XGBoost achieves the best performance, obtaining 97.8% accuracy, an F1-score of 0.976, and stable predictions across all classes. Random Forest ranks second with 96.4% accuracy, followed by Gradient Boosting at 95.7%. Meanwhile, Decision Tree and KNN exhibit lower performance due to their higher sensitivity to data variation. SHAP analysis reveals that the most influential features in phishing prediction include URL length, special character frequency, entropy levels, and the number of subdomains. These findings demonstrate that integrating XAI not only enhances model transparency but also ensures that phishing detection systems remain accurate, interpretable, and accountable.
Web-Based Real-Time Object Detection System with Audio Output for the Visually Impaired Muhammad Iqbal Fahrezzi; Azril Arfansyah; Augis Dinanti; Khairany Zuhriyyah Jinan Hsb; Hermawan Syahputra
KHARISMA Tech Vol 21 No 2 (2026): KHARISMATech Journal
Publisher : STMIK KHARISMA Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55645/kharismatech.v21i2.702

Abstract

Limited access to visual information is a major problem faced by visually impaired individuals, making it difficult to recognize surrounding objects and identify currency denominations independently. This issue highlights the need for a system capable of presenting visual information in a more accessible form. Therefore, this study aimed to develop a real-time object detection system based on a web platform with audio output to improve accessibility. The method included requirement analysis, system design, implementation using digital image processing and deep learning techniques, and system testing. The model applied a detection confidence threshold of ≥ 70% and achieved recognition accuracy of ≥ 85% under normal conditions. The system was also integrated with text-to-speech technology to deliver detection results in audio form. The results indicated that the system operated effectively in real-time with good responsiveness through a web browser without requiring additional installation. Therefore, the developed system proved capable of enhancing accessibility and supporting the independence of visually impaired users.
Interpretative Decision Tree Modeling for Identifying Depression Risk Factors in College Students Using PHQ-9 Data Munirah; Sunardi; Abdul Fadlil
KHARISMA Tech Vol 21 No 2 (2026): KHARISMATech Journal
Publisher : STMIK KHARISMA Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55645/kharismatech.v21i2.711

Abstract

This study aims to develop an interpretive Decision Tree model to identify depression levels in college students using Patient Health Questionnaire-9 (PHQ-9) data. Depression in college students is a growing mental health issue, while most machine learning models are still difficult to interpret. Therefore, this study uses Decision Tree as an explainable artificial intelligence (XAI) approach that is able to produce transparent and easy-to-understand decision rules. The research method uses a supervised learning approach with PHQ-9 data represented in text form and converted into numeric using Label Encoding. The target variable is formed based on the depression level category from the total PHQ-9 score. The model was evaluated using a 5-fold cross-validation technique with accuracy, precision, recall, and F1-score metrics. The results showed that the model achieved an accuracy of 54.4% in cross-validation and 68% in training data, with a macro F1-score of 0.67. Feature importance analysis showed that the PHQ2 variable was the most dominant factor in classifying depression. In addition, the Decision Tree structure was able to provide a clear interpretation of the pattern of depressive symptoms in college students. This study shows that Decision Tree has the potential to support early detection of student depression in a more transparent and interpretable manner.
PERANCANGAN APLIKASI SMART HYDROPONICS DALAM RUANGAN BERBASIS INTERNET OF THINGS MENGGUNAKAN METODE PROTOTYPING Bagus Muhammad Akbar; Abdul Fadlil; Sunardi
KHARISMA Tech Vol 21 No 2 (2026): KHARISMATech Journal
Publisher : STMIK KHARISMA Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55645/kharismatech.v21i2.714

Abstract

The growing population and limited land availability are driving the development of efficient alternative farming methods. Indoor hydroponics is an innovative soilless cultivation solution; however, manually managing environmental parameters such as pH, TDS/EC, temperature, humidity, CO₂, and tVOC is inefficient. This study designs the SHIRi (Smart Indoor Hydroponics) system based on the Internet of Things (IoT) using the prototyping method. The system utilizes an ESP32 microcontroller connected to pH, TDS/EC, DS18B20, DHT21, and SGP30 sensors, along with actuators including peristaltic pumps and a grow light. Data is transmitted via the MQTT protocol to a Mosquitto broker, stored in a Supabase database, and visualized through a responsive real-time web dashboard. The system supports two operating modes: Manual and Auto (rule engine). The prototyping method was applied iteratively in three cycles. Testing showed average pH sensor error of 0.1 units, TDS/EC 1.2ppm, water temperature 0.2°C, and humidity 1.5% so it has accuracy 97%. All dashboard features were successfully validated with an average update delay of 1.3 seconds. This system is expected to support autonomous and precision indoor hydroponics management.
SENTIMENT ANALYSIS OF USER REVIEWS OF THE SENTUH TANAHKU APPLICATION Kusmiarto; Wahyuni Wahyuni
KHARISMA Tech Vol 21 No 2 (2026): KHARISMATech Journal
Publisher : STMIK KHARISMA Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55645/kharismatech.v21i2.735

Abstract

Sentuh Tanahku is one of the digital land information service instruments initiated by the Ministry of Agrarian Affairs and Spatial Planning/National Land Agency in 2017. Since its launch, the application has been downloaded more than one million times and has received approximately 47,000 user ratings on the Google Play Store. This study examined 19,927 Google Play Store reviews from September 21, 2017 to May 13, 2026 to identify sentiment patterns, classification model performance, and the service aspects that most frequently influenced user evaluations. The analysis involved text preprocessing, rating-based labeling, sentiment lexicon analysis, TF-IDF representation, Multinomial Naive Bayes, Linear Support Vector Machine, and service-aspect mapping. The results showed a polarized rating pattern. Five-star reviews accounted for 57.93%, while one-star reviews reached 23.97%, with an average rating of 3.70. Lexicon analysis identified 9,605 positive reviews, 6,537 negative reviews, and 3,785 neutral reviews. Linear SVM produced the best macro F1-score of 62.29%. The most prominent complaints concerned account authentication, application performance, and the connectivity between land certificates and parcel data. These findings indicate that the quality of land service applications depends on technical stability, clear verification flows, and service data readiness.

Page 21 of 21 | Total Record : 206

 17   18   19   20   21 

Filter by Year

2018 2026


Reset
Filter By Issues
All Issue Vol 21 No 2 (2026): KHARISMATech Journal Vol 21 No 1 (2026): KHARISMATech Journal Vol 20 No 2 (2025): KHARISMATech Journal Vol 20 No 1 (2025): KHARISMATech Journal Vol 19 No 2 (2024): Jurnal KHARISMATech Vol 19 No 1 (2024): Jurnal KHARISMATech Vol 18 No 2 (2023): KHARISMATech Journal Vol 18 No 1 (2023): Jurnal KHARISMATech Vol 17 No 2 (2022): Jurnal KHARISMATech Vol 17 No 1 (2022): Jurnal KHARISMATEch Vol 16 No 2 (2021): Jurnal KHARISMA Tech Vol 16 No 1 (2021): Jurnal KHARISMA Tech Vol 15 No 2 (2020): KHARISMA Tech Journal Vol 14 No 2 (2019): Jurnal KHARISMA Tech Vol 14 No 1 (2019): Jurnal KHARISMA Tech Vol 13 No 2 (2018): Jurnal KHARISMATech Vol 13 No 1 (2018): Jurnal KHARISMA Tech More Issue