Garuda - Garba Rujukan Digital

SISTEM INFORMASI PEMANTAUAN STATUS GIZI BALITA Khasnur Hidjah; Helna Wardhana; Heroe Santoso; Anthony Anggrawan
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol 15 No 2 (2016)
Publisher : LPPM Universitas Bumigora

Based on interviews with staff nutrition Health Offce (Dikes) West Lombok, that is not currently available information systems that can be used to input data monitoring nutritional status of children. So it still takes a very long time to get the right information related to monitoring the nutritional status of children and families aware of nutrition per each district. The primary data sourced directly from the community gathered by Puskesmas offcers. Analysis of the data needed to meet the needs of data input, process and report to the monitoring system of nutritional status include: site identifcation, the identity of the household, the habit of weighing the family members, the question for pregnant or postpartum mothers, the nutritional intake of the family, the identity of a toddler, a child’s weight. The expected benefts of the outcomes defned as follows: enhance the ability to analyze the situation of food and nutrition in every region, able to set the priority handling of food and nutrition, able to monitor and evaluate the development of food and nutrition, improve community health status is marked as well as out of the category of problematic areas of health, especially malnutrition and less.

Expert System for Skin Disease Diagnosis Using the Best First Search Method and Fuzzy Tsukamoto Fahry, Fahry; Adam, M. Awaludin; Hidjah, Khasnur; Azwar, Muhammad; Hairani, Hairani
Journal of Electrical Engineering and Computer (JEECOM) Vol 7, No 1 (2025)
Publisher : Universitas Nurul Jadid

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33650/jeecom.v7i1.10981

The skin is the largest organ and is vulnerable to various diseases, which can spread through direct contact or the environment. Skin diseases are among the ten most common conditions in outpatient care in Indonesia, often caused by poor hygiene and environmental exposure. The limited number of dermatologists makes diagnosing and treating skin diseases more challenging. This study develops an expert system for diagnosing skin diseases using the Best First Search method and Fuzzy Tsukamoto, serving as an alternative or complement to medical diagnosis. Best First Search prioritizes diagnoses based on predefined rules, while Fuzzy Tsukamoto adds flexibility in assessing disease severity. Testing shows that the system achieves an accuracy of 83.3%, demonstrating its potential to assist patients and medical professionals in improving diagnostic efficiency and healthcare quality for skin diseases.

Classification of Learning Styles of Junior High School Students Using Random Forest & XGBoost Algorithm Christine Eirene; Dian Syafitri; Neny Sulistianingsih; Khasnur Hidjah; Hairani Hairani
Jurnal Bumigora Information Technology (BITe) Vol. 7 No. 1 (2025)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/bite.v7i1.4913

Background: Accurately identifying students' learning styles so that educators can adjust their teaching methods accordingly is a challenge in the field of education. However, the application of Machine Learning for learning style classification has not yet been implemented in schools in Mataram City. Objective: This study aims to classify the learning styles of students at Junior high school (SMP) Negeri 2 Mataram using Random Forest and XGBoost algorithms. Method: Data were collected through questionnaires completed by students in grades 7, 8, and 9. The results of data exploration (EDA) show data imbalance in the collected classes. Result: These results indicate that both algorithms performed well in classifying learning styles, with XGBoost showing slightly better performance. However, the accuracy obtained is not yet optimal, likely due to the limited dataset size. To address data imbalance, the SMOTE technique was applied. Initial evaluation showed that both XGBoost and Random Forest achieved an accuracy of 80%. After Hyperparameter Tuning, the accuracy of XGBoost increased to 84%, while Random Forest reached 82%. Conclusion: This study contributes to the application of Machine Learning in the education sector and highlights the need for further research to enhance model performance.

Feature Extraction in Eye Images Using Convolutional Neural Network to Determine Cataract Disease Fitra Rizki Ramdhani; Khasnur Hidjah; Muhammad Zulfikri; Hairani Hairani; Mayadi Mayadi; Ni Gusti ayu Dasriani; Juvinal Ximenes Guterres
International Journal of Engineering and Computer Science Applications (IJECSA) Vol. 4 No. 2 (2025): September 2025
Publisher : Universitas Bumigora Mataram-Lombok

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/ijecsa.v4i2.5064

The eye is one of the vital human senses and serves as the main organ for vision. One of the visual impairments that requires special attention is blindness, and cataracts are a major cause of it. A cataract is a condition in which the eye’s lens becomes cloudy due to changes in the lens fibers or materials inside the capsule. This cloudiness blocks light from entering the eye and reaching the retina, significantly interfering with vision. Early detection of cataracts is essential to prevent blindness. An efficient image-based classification model is needed for cataract detection. This study aims to test the Convolutional Neural Network (CNN) model for early cataract detection by exploring the use of several optimization algorithms: Adaptive Moment Estimation (Adam), Root Mean Square Propagation (RMSprop), Adaptive Gradient Algorithm (AdaGrad), and Stochastic Gradient Descent (SGD). The research method follows an experimental approach, where eye image datasets are trained using the same CNN architecture but with different parameter configurations. The results show that the Adam optimizer, with a data split of 70% for training, 15% for validation, and 15% for testing over 50 epochs, produced the best results, achieving accuracies of 94%, 93%, and 93%, respectively. Other optimizers performed reasonably well but could not match Adam's stability and accuracy. The implication of this research is that the choice of optimizer and hyperparameter configuration plays a crucial role in improving the performance of image-based cataract detection models.

Optimizing Sentiment Analysis for Lombok Tourism Using SMOTE and Chi-Square with Machine Learning Hairani; Anggrawan, Anthony; Muhammad Ridho Akbar; Khasnur Hidjah; Muhammad Innuddin
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 9 No 4 (2025): August 2025
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29207/resti.v9i4.6623

Tourism is a vital economic sector for Lombok Island, which is renowned for its natural beauty and cultural richness as a top destination. The rapid growth of tourism in Lombok requires a deep understanding of tourists' perceptions and sentiments to ensure an optimal service quality. The sentiment analysis of online reviews is valuable for identifying service strengths and weaknesses and addressing tourists' needs more effectively. This not only enhances tourist satisfaction, but also aids in the design of more effective marketing strategies. However, text data analysis from online reviews presents unique challenges such as noise, class imbalance, and numerous features that may affect classification results. Therefore, this study aims to classify tourist sentiment toward Lombok tourism using machine learning methods combined with feature selection and oversampling techniques. This study focuses on optimizing sentiment analysis of tourism-related tweets using a combination of SMOTE oversampling and Chi-Square feature selection on improving classification performance without hyperparameter tuning. The study applies machine learning methods, such as SVM and Naïve Bayes, with feature selection and oversampling using Chi-Square and SMOTE. The dataset used was sentiment data regarding Lombok tourism obtained from Twitter in 2023, consisting of 940 instances divided into three classes: Negative, Neutral, and Positive. The research findings show that the use of SMOTE and Chi-Square can improve the accuracy of the SVM and Naive Bayes methods. Without optimization, the SVM method achieved an accuracy of 73.93% and a Naive Bayes of 67.02%. After optimization with SMOTE and Chi-Square, the accuracy increased for SVM by 90% and Naive Bayes by 84% to classify tourist sentiment towards Lombok tourism. The implications indicate that combining data balancing using SMOTE with feature selection via Chi-Square effectively improves the performance of sentiment classification models for tourist opinions on Lombok's tourism.

Implementasi Konsultasi Stunting Balita Menggunakan Large Language Models (LLMs) Tanwir, Tanwir; Hidjah, Khasnur; Susilowati, Dyah
Reputasi: Jurnal Rekayasa Perangkat Lunak Vol. 6 No. 1 (2025): Mei 2025
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/reputasi.v6i1.8961

Stunting pada balita merupakan masalah kesehatan kritis di Indonesia yang memerlukan intervensi berbasis teknologi untuk meningkatkan akses informasi nutrisi. Penelitian ini bertujuan mengembangkan chatbot konsultasi stunting berbasis Large Language Models (LLMs) guna menyediakan rekomendasi kesehatan yang akurat dan mudah diakses. Metode yang digunakan berupa Model LLaMA 3 di-fine-tuning menggunakan dataset Q&A spesifik stunting berisi 7.642 entri, kemudian dievaluasi dengan matrik ROUGE untuk mengukur kesesuaian semantik respons. Hasil menunjukkan model Stunting mencapai skor ROUGE-1 (72,24%), ROUGE-2 (64,54%), ROUGE-L (70,42%), dan ROUGE-Lsum (70,96%), secara signifikan melampaui model baseline seperti LLaMA3, Deepseek-R1, dan Mistral. Chatbot diimplementasikan dalam aplikasi web berbasis cloud dengan arsitektur terdistribusi, dilengkapi enkripsi SSL dan HTTPS untuk menjamin keamanan data. Sistem ini memungkinkan interaksi real-time antara pengguna dan model LLMs melalui antarmuka berbasis Gradio. Temuan penelitian mengonfirmasi potensi LLMs dalam menyederhanakan layanan kesehatan preventif, khususnya di daerah dengan sumber daya terbatas

Hybrid Deep Learning Untuk Prediksi Kunjungan Tamu Hotel Satrani, Azral; Krismono, Bambang; Hidjah, Khasnur
Jurnal Sistem Informasi dan Teknologi Vol 5 No 2 (2025): Jurnal Sistem Informasi dan Teknologi (SINTEK)
Publisher : LPPM STMIK KUWERA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56995/sintek.v5i2.173

Prediksi jumlah kunjungan tamu hotel adalah aspek penting dalam pengelolaan operasional dan perencanaan strategis, terutama pasca pandemi Covid-19 yang menyebabkan fluktuasi tinggi dalam kunjungan. Holiday Resort Lombok, resort bintang empat di Senggigi, mencatat pertumbuhan kunjungan 35,20% dari 2022 hingga 2023, menunjukkan pemulihan pariwisata. Penelitian ini mengembangkan model prediksi menggunakan hybrid deep learning yang mengintegrasikan Convolutional Neural Network (CNN) untuk mengekstraksi pola spasial dan Long Short-Term Memory (LSTM) untuk menangani aspek temporal. Dataset terdiri dari 730 catatan harian kunjungan dari Januari 2022 hingga Desember 2023, dengan pelatihan model pada variasi epoch (50, 100, 150, dan 200). Hasil terbaik diperoleh pada 150 epoch, dengan Root Mean Sequare Error (RMSE) 29,55 untuk data pelatihan dan 32,23 untuk data pengujian, menunjukkan akurasi yang lebih baik dibandingkan metode tradisional. Namun, model menunjukkan potensi overfitting, memerlukan optimalisasi lebih lanjut. Model ini dapat mendukung pengambilan keputusan terkait alokasi sumber daya dan strategi pemasaran. Penelitian selanjutnya disarankan untuk mengeksplorasi ensemble learning dan integrasi variabel eksternal untuk meningkatkan ketepatan model.

Klasifikasi Bahasa Isyarat Huruf Hijaiyah Dengan GAN Dan CNN Cahaya Ardi, L. Nanda; Krismono, Bambang; Hidjah, Khasnur
Jurnal Sistem Informasi dan Teknologi Vol 5 No 2 (2025): Jurnal Sistem Informasi dan Teknologi (SINTEK)
Publisher : LPPM STMIK KUWERA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56995/sintek.v5i2.174

Pembelajaran huruf hijaiyah, termasuk untuk penyandang tunarungu yang menggunakan bahasa isyarat, sangat penting untuk memahami Al-Qur'an. Namun, pengembangan sistem klasifikasi berbasis kecerdasan buatan menghadapi kendala karena jumlah data citra bahasa isyarat huruf hijaiyah terbatas. Studi ini menyarankan metode hibrida yang menggunakan Generative Adversarial Network (GAN) untuk memperluas dataset dan Convolutional Neural Network (CNN) untuk mengklasifikasikan gambar bahasa isyarat huruf hijaiyah. Dataset awal, yang terdiri dari 1.877 gambar dengan 28 kelas yang dikumpulkan dari Roboflow, diperluas menggunakan DCGAN dan augmentasi gambar. Hasil pengujian menunjukkan bahwa integrasi GAN dan augmentasi dapat secara signifikan meningkatkan akurasi klasifikasi CNN, dengan tingkat akurasi tertinggi 95%. Metode ini terbukti efektif dalam meningkatkan kinerja sistem klasifikasi dan dapat digunakan sebagai media pembelajaran interaktif untuk penyandang disabilitas, terutama tunarungu.

Hybrid Deep Learning Untuk Prediksi Kunjungan Tamu Hotel Satrani, Azral; Krismono, Bambang; Hidjah, Khasnur
Jurnal Sistem Informasi dan Teknologi Vol 5 No 2 (2025): Jurnal Sistem Informasi dan Teknologi (SINTEK)
Publisher : LPPM STMIK KUWERA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56995/sintek.v5i2.173

Prediksi jumlah kunjungan tamu hotel adalah aspek penting dalam pengelolaan operasional dan perencanaan strategis, terutama pasca pandemi Covid-19 yang menyebabkan fluktuasi tinggi dalam kunjungan. Holiday Resort Lombok, resort bintang empat di Senggigi, mencatat pertumbuhan kunjungan 35,20% dari 2022 hingga 2023, menunjukkan pemulihan pariwisata. Penelitian ini mengembangkan model prediksi menggunakan hybrid deep learning yang mengintegrasikan Convolutional Neural Network (CNN) untuk mengekstraksi pola spasial dan Long Short-Term Memory (LSTM) untuk menangani aspek temporal. Dataset terdiri dari 730 catatan harian kunjungan dari Januari 2022 hingga Desember 2023, dengan pelatihan model pada variasi epoch (50, 100, 150, dan 200). Hasil terbaik diperoleh pada 150 epoch, dengan Root Mean Sequare Error (RMSE) 29,55 untuk data pelatihan dan 32,23 untuk data pengujian, menunjukkan akurasi yang lebih baik dibandingkan metode tradisional. Namun, model menunjukkan potensi overfitting, memerlukan optimalisasi lebih lanjut. Model ini dapat mendukung pengambilan keputusan terkait alokasi sumber daya dan strategi pemasaran. Penelitian selanjutnya disarankan untuk mengeksplorasi ensemble learning dan integrasi variabel eksternal untuk meningkatkan ketepatan model.

Klasifikasi Bahasa Isyarat Huruf Hijaiyah Dengan GAN Dan CNN Cahaya Ardi, L. Nanda; Krismono, Bambang; Hidjah, Khasnur
Jurnal Sistem Informasi dan Teknologi Vol 5 No 2 (2025): Jurnal Sistem Informasi dan Teknologi (SINTEK)
Publisher : LPPM STMIK KUWERA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56995/sintek.v5i2.174

Pembelajaran huruf hijaiyah, termasuk untuk penyandang tunarungu yang menggunakan bahasa isyarat, sangat penting untuk memahami Al-Qur'an. Namun, pengembangan sistem klasifikasi berbasis kecerdasan buatan menghadapi kendala karena jumlah data citra bahasa isyarat huruf hijaiyah terbatas. Studi ini menyarankan metode hibrida yang menggunakan Generative Adversarial Network (GAN) untuk memperluas dataset dan Convolutional Neural Network (CNN) untuk mengklasifikasikan gambar bahasa isyarat huruf hijaiyah. Dataset awal, yang terdiri dari 1.877 gambar dengan 28 kelas yang dikumpulkan dari Roboflow, diperluas menggunakan DCGAN dan augmentasi gambar. Hasil pengujian menunjukkan bahwa integrasi GAN dan augmentasi dapat secara signifikan meningkatkan akurasi klasifikasi CNN, dengan tingkat akurasi tertinggi 95%. Metode ini terbukti efektif dalam meningkatkan kinerja sistem klasifikasi dan dapat digunakan sebagai media pembelajaran interaktif untuk penyandang disabilitas, terutama tunarungu.

Title

Found 15 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 15 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 15 Documents
Search