Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Journal of Computer Networks, Architecture and High Performance Computing

APPLICATION OF K-NEAREST NEIGHBOUR, RECURSIVE ELIMINATION AND ADASYN ALGORITHMS ON DERMATITIS DISEASE CLASSIFICATION DATA Ramadhani, Daib Jidan; Siswa, Taghfirul Azhima Yoga; Pranoto, Wawan Joko
Journal of Computer Networks, Architecture and High Performance Computing Vol. 7 No. 3 (2025): Articles Research July 2025
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v7i3.6656

Abstract

Dermatitis is a common type of non-infectious skin disease frequently found in Indonesia. Its prevalence is influenced by several factors such as poor hygiene, environmental conditions, and climate change. Data from RSUD Jagakarsa recorded that from 1,066 skin disease cases between February 2023 and January 2024, approximately 62.2% were non-infectious, and 34.4% of those were classified as dermatitis. The diagnostic process for dermatitis is often challenging due to its symptom similarity with other skin conditions, leading to potential misclassification. Therefore, a more accurate and efficient classification approach is required to support medical professionals in identifying dermatitis cases effectively. This study proposes the use of a combination of machine learning methods: K-Nearest Neighbor (KNN) as the core classification algorithm, Recursive Feature Elimination (RFE) for feature selection, and Adaptive Synthetic Sampling (ADASYN) to handle class imbalance within the dataset. The data was sourced from UPTD Puskesmas Bontang Barat in 2024, consisting of 392 samples and 10 main features. Evaluation was conducted using a 10-fold cross-validation scheme. Results showed that the baseline KNN model achieved an average accuracy of 62.23%. With ADASYN applied, the accuracy improved to 63.56%, and further increased to 92.71% when combined with feature selection using RFE.
Implementation Of Decision Tree Algorithms For Classification Of Respiratory Infectious Diseases Fauzi; Taghfirul Azhima Yoga Siswa; Fendy Yulianto
Journal of Computer Networks, Architecture and High Performance Computing Vol. 7 No. 4 (2025): Articles Research October 2025
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v7i4.6956

Abstract

Acute Respiratory Infection (ARI) is a common respiratory illness that frequently affects children, primarily caused by viruses such as rhinovirus or adenovirus. In Indonesia, a total of 200,000 ARI cases were recorded during the 2021–2023 period. This study aims to implement the Decision Tree algorithm to classify ARI cases. The dataset consists of 1,501 patient records obtained from UPT Puskesmas Bontang Barat for the 2024–2025 period. The research process includes the pre-processing stage, data splitting into training and testing sets using the 10-Fold Cross Validation technique. Subsequently, model evaluation is conducted using the Confusion Matrix to calculate the Accuracy, Precision, Recall, and F1-Score metrics. The results show that the Decision Tree algorithm is capable of performing classification with good performance, achieving an average accuracy of 81.75%, precision of 79.58%, recall of 81.75%, and an F1-score of 80.45%.