This research is a comparison using 3 (three) algorithms, namely Lo- gistic Regression, K-Nearest Neighbor, and Support Vector Machine in senti- ment analysis about the JMO application, as the main means for participants in the employment social security program, which plays a crucial role in providing services that meet participants' needs well. This research aims to compare three different classification algorithms for sentiment analysis in the Jamsostek mobile application. The process involves several stages, including data collection (Crawling), word separation (tokenizing), normalization, removal of common words (Stopword), and word simplification (Stemming). After the processing stage, the data is labeled and classified using a comparison of three algorithms. The results of the 3 tweet category algorithms tend to be positive and negative. From the Logistic Regression algorithm, the accuracy level achieved was 84.78%, the precision was 87.24%, and the recall was 62.16%, then the Support Vector Machine algorithm achieved an accuracy level was 89.13%, the precision was 86.67%, and the recall of 76.88%, and the KNN algorithm produced an ac- curacy level of 88.59%, precision of 91.07%, and recall of 71.88%.
7 September 2024 in Building of Informatics, Technology and Science (BITS)
Cardiovascular disease is a disease with a fairly high number of deaths. In Indonesia, the term cardiovascular is more popular with heart disease, which is a condition that can cause narrowing and blockage of blood vessels. Cardiovascular disease has two risks, the first is a risk that can be changed, such as stress, increased blood pressure, unhealthy diet, increased glucose levels, abnormal cholesterol and lack of physical activity. Meanwhile, risks that cannot be changed include family disease, gender, age and obesity. In this research, we can examine and analyze the performance of the two best classification algorithms, namely the decision tree algorithm and the random forest algorithm, in classifying cardiovascular disease based on the cause of the disease. The aspects studied are the performance results of each algorithm and evaluated using Area Under the Curve (AUC), classification report, k-Fold Cross Validation and Confusion matrix. The dataset used was taken from the Kaggle website with the data used being Cardiovascular Disease data which consists of 68.205 rows (patient data) and 17 attributes. . Based on the evaluation results using the Area Under The Curve (AUC) value, the highest result was obtained at 0.761 by the Random Forest algorithm with balanced data conditions with Random oversampling. Meanwhile, the lowest AUC value was obtained by the Decision Tree algorithm with unbalanced data of 0.592. Based on these results, it is known that the Random Forest algorithm with a balanced data scheme is a better algorithm, with a balanced data scenario using SMOTE and Random Oversampling techniques.
5 June 2025 in JOURNAL OF APPLIED INFORMATICS AND COMPUTING
This research aims to analyze public sentiment toward the boycott movement against Israel on the X platform by applying Random Forest and Logistic Regression algorithms. The study uses 616 tweets collected through web crawling with relevant keywords such as "Boikot", "Israel", and "Palestine", covering the period from March 1, 2023 to January 30, 2025. The dataset underwent preprocessing including cleaning, normalization, stopword removal, tokenization, and stemming. Sentiment labeling was conducted both manually, categorizing the data into positive, negative, and neutral classes. TF-IDF was used for feature weighting. The data was split into 80% training and 20% testing. The Random Forest model achieved an accuracy of 70%, while Logistic Regression reached 68%. Both models showed higher accuracy in predicting positive sentiment compared to negative and neutral. The results suggest that public opinion on the boycott movement on social media tends to be supportive, with “Boikot,” “Israel,” and “Palestine” being the most dominant terms. Random Forest performed slightly better in classification, though improvements are needed in recognizing non-positive sentiments.
Sistem deteksi kendaraan bermotor yang digunakan pada sistem E-TLE bukan hal baru, dikarenakan sistem tersebut sudah banyak diteliti dan diimplementasikan dinegara maju. Deteksi yang dimaksud adalah mendeteksi nomor polisi yang ada pada suatu kendaraan kemudian dikonversi menjadi laporan dan dikategorikan dalam bentuk pelanggaran seperti apa sesuai perundang-undangan. Namu, tidak banyak Masyarakat yang paham akan alur proses deteksi dan akurasi dari sistem tersebut. Sehingga dibutuhkan sosialisasi kepada Masyarakat lebih lanjut dan mendalam. Sosialisasi dilakukan pada kegiatan pengabdian kepada masyakarat yang dilakukan ditingkat desa yang berada dibagian luar Kabupaten Karawang. Masyarakat yang mengikuti adalah perangkat desa setingkat RT dikarenakan nanti akan disosialisasikan kembali oleh aparatur desa kepada Masyarakat. Peserta berjumlah 30 orang yang dibagi menjadi 2 termin pagi dan sore, dikarenakan kapasitas ruangan dan faktor efektifitas penjelasan agar kondusif. Hasilnya 50% yaitu berjumlah 15 orang tidak mengetahui perihal sistem seperti ini bisa akurat dan mengidentifikasikan nomor kendaraan serta pelaggarannya
26 May 2025 in Jurnal Sisfokom (Sistem Informasi dan Komputer)
Road damage is one of the leading factors contributing to traffic accidents. Rapid identification and repair of damaged roads are crucial in road infrastructure management. This study aims to develop an effective method for detecting road damage, utilizing the YOLOv9 algorithm as a key component, such as cracks and potholes, using the Convolutional Neural Network (CNN) approach. YOLOv9 was chosen due to its efficient architecture, which enables real-time object detection, and its proven effectiveness in various object detection tasks. An annotated dataset of road images was used during the model training and testing process. The results show that the YOLOv9 model can accurately detect road damage. The model achieved a precision of 0.85 and a recall of 0.992 for pothole detection, and a precision of 0.94 for crack detection. Evaluation using mAP50 yielded a score of 0.96, while mAP50-95 reached 0.77, indicating strong detection and classification capability. A consistent decline in loss functions during training also signifies effective learning by the model. These findings suggest that YOLOv9 has the potential to be implemented in automated road damage detection systems, which can accelerate maintenance processes and enhance road user safety.