Article Per Year (5 Year)
p-Index From 2021 - 2026
0.562
P-Index
This Author published in this journals
All Journal Journal of Artificial Intelligence and Digital Business Blueprint Journal
Chrisnawati, Giantika
Unknown Affiliation
Civil Engineering, Building, Construction & Architecture Computer Science & IT Economics, Econometrics & Finance Electrical & Electronics Engineering Engineering Industrial & Manufacturing Engineering

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
Analisis Perbandingan Algoritma K-Nearest Neighbor dan Ensemble Learning dalam Klasifikasi Penyakit Obesitas Tondang, Sarihot; Prasetyo, Ramadhan Roy; Fulvian, Rafi; Sitorus, Yosua Goldstein; Chrisnawati, Giantika
RIGGS: Journal of Artificial Intelligence and Digital Business Vol. 4 No. 2 (2025): Mei - Juli
Publisher : Prodi Bisnis Digital Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/riggs.v4i2.994

Abstract

Penelitian ini bertujuan untuk membandingkan performa tiga algoritma machine learning, yaitu K-Nearest Neighbors (KNN), Random Forest, dan Gradient Boosting, dalam tugas klasifikasi tingkat obesitas berdasarkan dataset UCI “Estimation of Obesity Levels”. Dataset ini terdiri dari 2111 sampel dengan 17 atribut, termasuk fitur numerik dan kategorikal, serta label klasifikasi “NObeyesdad” dengan tujuh kelas obesitas. Proses pra-pemrosesan data melibatkan normalisasi menggunakan StandardScaler untuk fitur numerik, one-hot encoding untuk fitur kategorikal, dan Synthetic Minority Oversampling Technique (SMOTE) untuk mengatasi ketidakseimbangan kelas. Model-model dilatih menggunakan pipeline yang mencakup pra-pemrosesan dan klasifikasi, dengan optimasi hyperparameter melalui GridSearchCV dan validasi silang 5-fold. Evaluasi dilakukan dengan metrik akurasi, precision, recall, F1-score, dan analisis confusion matrix. Hasil menunjukkan bahwa Random Forest mencapai performa tertinggi dengan akurasi 98.6%, diikuti oleh Gradient Boosting dengan akurasi 98.1%, dan KNN dengan akurasi 86.8%. Random Forest menunjukkan stabilitas prediksi yang superior, terutama pada kelas-kelas dengan fitur serupa, sementara Gradient Boosting juga menawarkan performa yang konsisten. KNN, meskipun sederhana, cenderung kurang stabil dalam menangani data multi-kelas yang kompleks. Penelitian ini memberikan wawasan penting mengenai penerapan algoritma machine learning dalam diagnosis obesitas, dengan Random Forest sebagai pilihan terbaik untuk klasifikasi akurat dan stabil.
Rw Segmentation Analysis for the Climate Village Program as a Basis for Planning in South Jakarta Using K-Means Clustering Marni Berek, Maria Susey; Taufiq, Ghofar; Chrisnawati, Giantika
Blueprint Journal Vol 1 No 2 (2025): Agustus: Blueprint Journal
Publisher : PT Yupin Felicitas Utama

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

Abstract

Climate change is a global issue that has multidimensional impacts on human life, including in Indonesia. In response to this challenge, the government developed the Climate Village Program (PROKLIM) which prioritizes community empowerment through a community-based approach. This program aims to strengthen climate change adaptation and mitigation efforts through participatory local resource management. This study uses the K-Means clustering method to group areas based on environmental characteristics at the Neighborhood Association (RW) level, in order to identify patterns and support decision making in effective environmental management. This study proves that the K-Means Clustering method is effective in grouping RWs in South Jakarta based on indicators relevant to the Climate Village Program (ProKlim). The latest report from the World Meteorological Organization (2024) states that 2023 was the hottest year in history, with an anomaly (Hasbullah & Assyahri, 2025) of global temperatures reaching 1.45°C above the average temperature in the pre-industrial era. Furthermore, the last nine years (2015–2023) were recorded as the period with the hottest consecutive temperatures in the history of climate records. The segmentation results show clear differences between groups in terms of levels of vulnerability to climate change, community engagement, and environmental preparedness. This grouping provides a strong, data-driven analytical basis, allowing the South Jakarta Environmental Agency (DLH) to use it as a strategic reference for more targeted and targeted planning and implementation of ProKlim. 
Analysis of Reading Interests of Visitors to the Library of State Junior High School 01 Salem Using the K-Means Clustering Algorithm Nurhotimah, Ica; Taufiq, Ghofar; Chrisnawati, Giantika
Blueprint Journal Vol 1 No 2 (2025): Agustus: Blueprint Journal
Publisher : PT Yupin Felicitas Utama

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

Abstract

Advances in information and communication technology have made information an important element in people's lives, including education and literacy. The high demand for information has encouraged the application of data mining techniques that combine statistics, mathematics, artificial intelligence, and machine learning to extract useful information from big data. In the context of libraries, data mining can be used to analyze borrowing and visit data to understand user needs patterns. In Indonesia, low reading interest remains a serious issue. Data from PISA (OECD) and UNESCO reports indicate that Indonesia's literacy skills and reading interest levels are below global standards. Many students are not accustomed to accessing reading materials outside of textbooks and rarely visit school libraries, which should serve as centers for literacy. The grouping was based on features such as student names, class, book titles along with publishers and authors, and the date and time of visits to the library. This data was categorized into three groups: high, moderate, and low reading interest. The clustering results using the K-Means Clustering algorithm at SMP Negeri 01 Salem show that the majority of students (118) fall into the low reading interest category, 13 students into the moderate category, and only 1 student (Dian) into the high reading interest category. Evaluating the quality of the clusters using the Davies–Bouldin Index (DBI) yielded a value of 0.2966, indicating very good cluster quality—a low DBI value indicates compact and clearly separated clusters. These results prove that the K-Means algorithm is effective in grouping students based on reading behavior. With this segmentation, schools can develop data-driven literacy strategies: tailoring book collections to each cluster's preferences, conducting special programs for students with low reading interest, and involving students with high reading interest as literacy ambassadors. This approach is expected to increase student engagement and strengthen the reading culture at school.
Co-Authors Fulvian, Rafi Ghofar Taufiq, Ghofar Marni Berek, Maria Susey Nurhotimah, Ica Prasetyo, Ramadhan Roy Sitorus, Yosua Goldstein Tondang, Sarihot