Claim Missing Document
Check
Articles

Found 16 Documents
Search

Analysis Of Public Sentiment Towards The Free Nutritious Meal Program In Schools Based On Tweets Using The K-Nearest Neighbors Method Maharani, Aiga Rizki; Gustriansyah, Rendra; irfani, muhammad haviz
JATISI Vol 12 No 4 (2025): JATISI (Jurnal Teknik Informatika dan Sistem Informasi)
Publisher : Universitas Multi Data Palembang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35957/jatisi.v12i4.13446

Abstract

The public sentiment analysis of the free nutritious meal program in schools was conducted based on data from the social media platform Twitter (X). This program is an initiative by the Indonesian government aimed at improving the nutritional quality of children, particularly those from underprivileged families, as well as reducing stunting rates. The data used consisted of 3,007 tweets that had undergone preprocessing, manual labeling, and class balancing using oversampling techniques. The K-Nearest Neighbors (K-NN) method was applied to classify sentiment into three categories: positive, negative, and neutral. The data was split with 80% used for training and 20% for testing. The analysis process included data representation using TF-IDF and model evaluation using metrics such as accuracy, precision, recall, and F1-score. Evaluation results showed that the K-NN model with K=3 achieved an accuracy of 82%, with the best performance in classifying negative sentiment tweets (recall = 1.00, F1-score = 0.93). These findings indicate that public opinion toward the program tends to be negative, mainly due to concerns over budget allocation and food distribution. This study is expected to provide input for the government in designing more effective and responsive communication strategies and public policies.
INFLUENCE OF LEAF IMAGING DISTANCE ON WATER GUAVA CLASSIFICATION USING NEURAL NETWORK WITH GRAY LEVEL CO-OCCURRENCE MATRIX FEATURES Muhammad Haviz Irfani; Gasim; Andika Afrianto
Jurnal Riset Informatika Vol. 8 No. 1 (2025): Desember 2025
Publisher : Kresnamedia Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34288/jri.v8i1.419

Abstract

The development of Computer Vision technology has made a significant contribution to the agricultural sector, particularly in the identification of plants based on visual characteristics. Water guava (Syzygium aqueum) is one of the fruit commodities widely cultivated in Indonesia; however, its seedling varieties are often difficult to distinguish visually. Conventional methods relying on human observation tend to have low accuracy, highlighting the need for an accurate and efficient identification system from the early stages. This study aims to analyze the effect of varying imaging distances on the extraction results of leaf vein texture features using the Gray Level Co-occurrence Matrix (GLCM) method and to evaluate how this parameter influences the classification performance of water guava seedlings using the Backpropagation Artificial Neural Network (ANN). Unlike previous GLCM–ANN plant classification studies that primarily focused on lighting or species variation, this work systematically investigates imaging distance as a key factor in optimizing texture feature stability and improving model accuracy. Experiments were conducted using five imaging distances—7 cm, 9 cm, 11 cm, 13 cm, and 15 cm—with 2,500 images used for training data and 500 images for testing data. The results show that an imaging distance of 13 cm yielded the best performance, achieving 80% accuracy, where 80 out of 100 test images were correctly classified, supported by balanced precision, recall, and F1-score values indicating stable and reliable classification performance.
Pengaruh Deteksi Tepi Citra Urat Daun Pada Pengenalan Jenis Bibit Jeruk Menggunakan Metode Pengenalan JST-PB dan GLCM Dimas Apriandi; Gasim; Muhammad Haviz Irfani; Muhammad Ikhwan Jambak
Jurnal Software Engineering and Computational Intelligence Vol 3 No 02 (2025)
Publisher : Informatics Engineering, Faculty of Computer Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36982/jseci.v3i02.6219

Abstract

Identifikasi awal jenis bibit jeruk sangat penting untuk menjamin kualitas bibit dan meningkatkan produktivitas pertanian. Identifikasi secara manual membutuhkan keahlian khusus dan rentan terhadap kesalahan. Penelitian ini bertujuan untuk menganalisis pengaruh metode deteksi tepi terhadap akurasi klasifikasi bibit jeruk menggunakan Jaringan Syaraf Tiruan Propagasi Balik (JST-PB) dan fitur tekstur yang diekstraksi dengan Gray Level Co-occurrence Matrix (GLCM). Tiga metode deteksi tepi yaitu Canny, Laplacian of Gaussian (LoG), dan Roberts yang diterapkan pada citra daun jeruk dari empat varietas antara lain: Kunci, Nipis, Purut, dan Sambal. Fitur tekstur berupa contrast, correlation, homogeneity, dan entropy digunakan sebagai masukan dalam pelatihan JST-PB. Hasil penelitian menunjukkan bahwa metode Roberts dengan 30 neuron tersembunyi memberikan kinerja terbaik dengan precision rata-rata 75,56%, recall 75,00%, dan F1-score 74,97%. Hal ini menunjukkan bahwa pemilihan metode deteksi tepi berpengaruh signifikan terhadap akurasi klasifikasi. Kombinasi metode deteksi tepi Roberts, ekstraksi fitur GLCM, dan JST-PB terbukti efektif untuk pengenalan otomatis jenis bibit jeruk berbasis citra digital.    
Implementasi YOLO Framework pada Deteksi Otomatis Sepatu Badminton Pemain di Lapangan Ida Bagus Bisma; Rudi Heriansyah; Muhammad Haviz Irfani
Jurnal Komputer dan Teknik Informatika Vol. 1 No. 2 (2026): Edisi: Februari-April
Publisher : Pustaka Bangsa Indonesia

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

Abstract

Penggunaan sepatu yang tidak sesuai saat bermain badminton dapat meningkatkan risiko cedera serta merusak flooring lapangan. Oleh karena itu, diperlukan sistem deteksi otomatis untuk mengidentifikasi jenis sepatu yang digunakan pemain guna meningkatkan kepatuhan terhadap peraturan lapangan. Penelitian ini mengusulkan sistem deteksi sepatu badminton menggunakan algoritma Convolutional Neural Network (CNN) dengan framework You Only Look Once (YOLO). Dataset yang digunakan terdiri dari tiga kelas, yaitu Badminton Shoes, Warning Shoes, dan Foot, yang diperoleh dari situs resmi produk sepatu serta platform Roboflow. Proses pelatihan model meliputi tahap anotasi, preprocessing, dan augmentasi data. Evaluasi kinerja model dilakukan menggunakan confusion matrix dengan metrik accuracy, precision, recall, dan F1-score. Hasil penelitian menunjukkan bahwa model mampu mendeteksi dan mengklasifikasikan jenis sepatu dengan tingkat akurasi yang tinggi. Sistem ini berpotensi membantu pengelola lapangan dalam melakukan pengawasan otomatis serta mengurangi risiko cedera akibat penggunaan sepatu yang tidak sesuai.
Pengaruh Splitting Data terhadap Akurasi Klasifikasi Demam Berdarah Dengue Menggunakan K-Nearest Neighbors Ibrahim Akbar, Muhammad; Rudi, Rudy Heriansyah; irfani, muhammad haviz
Algoritme Jurnal Mahasiswa Teknik Informatika Vol 6 No 2 (2026): April 2026 || Algoritme Jurnal Mahasiswa Teknik Informatika
Publisher : Program Studi Teknik Informatika Universitas Multi Data Palembang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35957/algoritme.v6i2.13274

Abstract

Early detection of Dengue Hemorrhagic Fever (DHF) is crucial to prevent serious complications and improve treatment effectiveness, particularly in high-case areas such as the Dempo Primary Health Center. This study aims to develop and evaluate a DHF classification system using the K-Nearest Neighbors (K-NN) algorithm with an optimal K value of 5, determined via the Elbow Method. The dataset consists of 200 medical records with an imbalanced class distribution between positive and negative DHF cases. Three data-splitting scenarios (70:30, 80:20, and 90:10) were tested to analyze the effect of data proportion on model performance. Evaluation metrics included accuracy, precision, recall, and F1-score. Results show that the 70:30 scenario achieved the best performance, with 90% accuracy, 96.67% precision, 85.29% recall, and 90.62% F1-score. For comparison, K-NN was tested against Decision Tree and Support Vector Machine (SVM) algorithms as baselines. K-NN demonstrated competitive and more stable performance, with an average accuracy difference of ±2% compared to the other methods. These findings confirm that K-NN provides reliable results for medical data with limited sample size and imbalanced class distribution. This study contributes empirical analysis regarding the influence of varying data split ratios on classification model stability and strengthens the application of machine learning for early DHF detection based on local medical data.
Analisis Struktur Tulisan Tangan melalui Deteksi Zona Spasi Antarkata Menggunakan CNN Mendalam Mufti, Nabilah; Heriansyah, Rudi; Irfani, Muhammad Haviz
SMARTICS Journal Vol 12 No 1 (2026): Journal SMARTICS (April 2026)
Publisher : Universitas PGRI Kanjuruhan Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21067/smartics.v12i1.13937

Abstract

Advances in image processing and deep learning technology enable more accurate handwriting analysis, including the detection of interword spacing, which exhibits high complexity due to variations in writing styles. This study aims to implement a Convolutional Neural Network (CNN) algorithm using the You Only Look Once version 11 (YOLOv11) framework to detect and classify interword spacing zones into three classes: Narrow Word Spacing (NWS), Medium Word Spacing (MWS), and Wide Word Spacing (WWS). The dataset comprises 150 handwritten images with a total of 4.117 annotated interword spacing objects. The research methodology involves testing the model across variations of learning rates (0.1, 0.01, 0.001, and 0.0001) and data split ratios (70:30, 80:20, and 90:10). Model performance was evaluated using Precision, Recall, F1-Score, and mean Average Precision (mAP) metrics. Based on 12 experimental trials, the best configuration was achieved with a learning rate of 0.001 and a 90:10 data split. This configuration produced an mAP@50 of 0.455, an mAP@50–95 of 0.261, and an F1-Score of 0.49. These results indicate that the YOLOv11 model is capable of detecting interword spacing zones with reasonably good performance, despite remaining classification errors due to visual similarities between classes.