Claim Missing Document
Check
Articles

Found 32 Documents
Search

PERBANDINGAN ALGORITMA SUPPORT VECTOR MACHINE (SVM) DAN NAÏVE BAYES DALAM KLASIFIKASI PENYAKIT DIABETES Desiani, Anita; Dewi, Novi Rustiana; Arhami, Muhammad; Sitorus, Dina Suzzete; Rahmadita, Suristhia
POSITIF : Jurnal Sistem dan Teknologi Informasi Vol 10 No 1 (2024): Positif : Jurnal Sistem dan Teknologi Informasi
Publisher : P3M Politeknik Negeri Banjarmasin

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31961/positif.v10i1.2092

Abstract

High levels of sugar in the blood can cause diabetes. The longer people are unable to control glucose in their blood, the more complications it can cause, other diseases and even death. Early detection of diabetes is needed, one way is by carrying out data mining classification. Data mining classification in this research uses two algorithms, namely SVM (Support Vector Machine) and Naïve Bayes. This research compares the two algorithms using two methods, namely training split and k-fold cross validation which aims to get the best classification results in detecting diabetes. The best classification results are determined by calculating the average value of precision, recall and accuracy. Based on this research, the SVM algorithm with split percentage training produces average values for precision, recall and accuracy, namely 77%, 71.5%, 77.27%, while the SVM algorithm with k-fold cross validation produces average values for precision, recall , and accuracy is 77%, 72.5%, 71%. The Naïve Bayes algorithm with the split percentage training method produces average values for precision, recall and accuracy, namely 75.5%, 74.5%, 79%, while the Naïve Bayes algorithm with k-fold cross validation produces average values for precision, recall, and accuracy of 75.5%, 74.5%, 75%. The best classification result in detecting diabetes is the Naïve Bayes algorithm, the split percentage method, which provides the best accuracy, precision and recall values above 74%.
Improve of Multiobjective Model on the Classification Problem of Food Consumption Levels in Indonesia Susanti, Eka; Dewi, Novi Rustiana; Arsi, Arsi
KUBIK Vol 10, No 1 (2025): KUBIK: Jurnal Publikasi Ilmiah Matematika
Publisher : Jurusan Matematika, Fakultas Sains dan Teknologi, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/kubik.v10i1.40632

Abstract

Classification is the process of grouping objects based on similarities and differences. In this article, a multi-objective classification model is developed with three objective functions, namely the function that maximizes the values of accuracy, sensitivity and specificity. The developed model is applied to the problem of classifying meat, egg and fish consumption levels. The classification method used is K-Nearest Neighbor (KNN) with three objective functions and the addition of the GridSearchCV module to the KNN calculation. Completion of the multiobjective model using the weighting method and Particle Swam Optimization (PSO). Based on the data, with objective function weights of 1, 2 and 3 respectively being 0.7, 0.15 and 0.15, the results obtained for Rural Areas Meat, Fish and Egg Attributes of the model performance are in good criteria. for Urban Areas Attributes of Meat, Fish and Eggs the model's performance in the criteria is very good. Addition of the GridsearchCV module can facilitate the calculation of the KNN method classification because the model will provide the best k value without having to do repeated calculations.
Improve of Multiobjective Model on the Classification Problem of Food Consumption Levels in Indonesia Susanti, Eka; Dewi, Novi Rustiana; Arsi, Arsi
KUBIK Vol 10 No 1 (2025): KUBIK: Jurnal Publikasi Ilmiah Matematika
Publisher : Department of Mathematics, Faculty of Science and Technology, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/kubik.v10i1.40632

Abstract

Classification is the process of grouping objects based on similarities and differences. In this article, a multi-objective classification model is developed with three objective functions, namely the function that maximizes the values of accuracy, sensitivity and specificity. The developed model is applied to the problem of classifying meat, egg and fish consumption levels. The classification method used is K-Nearest Neighbor (KNN) with three objective functions and the addition of the GridSearchCV module to the KNN calculation. Completion of the multiobjective model using the weighting method and Particle Swam Optimization (PSO). Based on the data, with objective function weights of 1, 2 and 3 respectively being 0.7, 0.15 and 0.15, the results obtained for Rural Areas Meat, Fish and Egg Attributes of the model performance are in good criteria. for Urban Areas Attributes of Meat, Fish and Eggs the model's performance in the criteria is very good. Addition of the GridsearchCV module can facilitate the calculation of the KNN method classification because the model will provide the best k value without having to do repeated calculations.
Improve of Multiobjective Model on the Classification Problem of Food Consumption Levels in Indonesia Susanti, Eka; Dewi, Novi Rustiana; Arsi, Arsi
KUBIK Vol 10 No 1 (2025): IN PRESS
Publisher : Jurusan Matematika, Fakultas Sains dan Teknologi, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/kubik.v10i1.40632

Abstract

Classification is the process of grouping objects based on similarities and differences. In this article, a multi-objective classification model is developed with three objective functions, namely the function that maximizes the values of accuracy, sensitivity and specificity. The developed model is applied to the problem of classifying meat, egg and fish consumption levels. The classification method used is K-Nearest Neighbor (KNN) with three objective functions and the addition of the GridSearchCV module to the KNN calculation. Completion of the multiobjective model using the weighting method and Particle Swam Optimization (PSO). Based on the data, with objective function weights of 1, 2 and 3 respectively being 0.7, 0.15 and 0.15, the results obtained for Rural Areas Meat, Fish and Egg Attributes of the model performance are in good criteria. for Urban Areas Attributes of Meat, Fish and Eggs the model's performance in the criteria is very good. Addition of the GridsearchCV module can facilitate the calculation of the KNN method classification because the model will provide the best k value without having to do repeated calculations.
Penerapan Konsep Bangun Fraktal pada Pembelajaran Bilangan Berpangkat Dewi, Novi Rustiana; Susanti, Eka; Sukanda, Dian Cahyawati; Dwipurwani, Oki; Zayanti, Des Alwine
Abdimas Galuh Vol 7, No 1 (2025): Maret 2025
Publisher : Universitas Galuh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25157/ag.v7i1.16890

Abstract

Kegiatan Pengabdian kepada Masyarakat ini bertujuan meningkatkan pemahaman siswa SD terhadap konsep bilangan berpangkat melalui visualisasi bangun fraktal Sierpinski menggunakan aplikasi Wolfram Alpha. Kegiatan dilaksanakan di SD Negeri 60 Palembang dengan peserta 52 siswa kelas 5 dan 6 serta 17 guru. Metode kegiatan adalah pengenalan aplikasi dan pendampingan belajar. Hasil evaluasi menunjukkan peningkatan pemahaman siswa kelas 6 dari rata-rata 35% pada pre-test menjadi 65% pada post-test. Sedangkan untuk kelas 5 meningkat dari 48% menjadi 52%. Pendampingan menggunakan Wolfram Alpha kepada guru dapat meningkatkan keterampilan dalam membuat media pembelajaran yang bervariasi. Hasil kegiatan pengabdian menunjukkan bahwa pendekatan pembelajaran berbasis teknologi melalui penerapan bangun fraktal Sierspinski yang divisualisasikan menggunakan aplikasi Wolfram Alpha memberikan dampak positif.
Implementasi Teknik Ensemble Stacking pada Klasifikasi Penyakit Anemia anita, anita desiani; Mukhlisah, Nur; Indira, Ria; Novi Rustiana Dewi; Yuli Andriani
JIEET (Journal of Information Engineering and Educational Technology) Vol. 9 No. 1 (2025)
Publisher : Universitas Negeri Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/jieet.v9n1.p51-55

Abstract

Abstrak— Anemia adalah penyakit yang disebabkan oleh kondisi seseorang yang memiliki kadar hemoglobin (Hb) darah dibawah normal. Deteksi penyakit dapat menggunakan bantuan data mining untuk mengklasifikasikan penyakit. Algoritma yang digunakan dalam penelitian ini adalah Gaussian Naive Bayes, K-Nearest Neighbor dan Support Vector Machine yang kemudian diterapkan pada teknik ensemble stacking. Penerapan Ensemble bertujuan untuk mendapatkan nilai keakurasian yang lebih baik dari klasifikasi individu. Pengujian algoritma ini menggunakan dua teknik pengujian yaitu percentage split dan k-fold cross validation. Untuk percentage split menggunakan ukuran split sebesar 80% data training dan 20% data uji dan pada k-fold cross validation dipilih nilai k=10. Hasil klasifikasi dari algoritma-algoritma tersebut memperoleh bahwa percentage split mendapatkan hasil akurasi yang lebih baik dibandingkan k-fold cross validation. Algoritma Support Vector Machine (SVM), Gaussian Naive Bayes dan k-Nearest Neighbor (kNN) dengan teknik pengujian percentage split memperoleh hasil akurasi secara berturut-turut sebesar 90,16%, 94,61% dan 96,49%. K-Nearest Neighbor (kNN) menghasilkan nilai akurasi tertinggi dari ketiga algoritma tersebut, namun dengan penerapan teknik ensemble memberikan kenaikan akurasi sebesar 1.05% dari hasil k-Nearest Neighbor (kNN). Ensemble dengan model stacking memperoleh hasil akurasi sebesar 97,19%. Berdasarkan hasil yang diperoleh dapat disimpulkan bahwa ensemble dengan model stacking dengan teknik pengujian percentage split memperoleh kinerja yang terbaik dari algoritma lainnya pada klasifikasi penyakit anemnia. Kata Kunci— Ensemble Learning, Support Vector Machine, Gaussian Naive Bayes, K-Nearest Neighbor, Anemia
OPTIMIZATION OF RICE INVENTORY USING FUZZY INVENTORY MODEL AND LAGRANGE INTERPOLATION METHOD Susanti, Eka; Puspita, Fitri Maya; Yuliza, Evi; Supadi, Siti Suzlin; Dwipurwani, Oki; Dewi, Novi Rustiana; Ramadhan, Ahmad Farhan; Rindarto, Ahmad
BAREKENG: Jurnal Ilmu Matematika dan Terapan Vol 17 No 3 (2023): BAREKENG: Journal of Mathematics and Its Applications
Publisher : PATTIMURA UNIVERSITY

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30598/barekengvol17iss3pp1215-1220

Abstract

Interpolation is a method to determine the value that is between two values and is known from the data. In some cases, the data obtained is incomplete due to limitations in data collection. Interpolation techniques can be used to obtain approximate data. In this study, the Lagrange interpolation method of degree 2 and degree 3 is used to interpolate the data on rice demand. A trapezoidal fuzzy number expresses the demand data obtained from the interpolation. The other parameters are obtained from company data related to rice supplies and are expressed as trapezoidal fuzzy numbers. The interpolation accuracy rate is calculated using Mean Error Percentage (MAPE). The second-degree interpolation method produces a MAPE value of 30.76 percent, while the third-degree interpolation has a MAPE of 32.92 percent. The quantity of order respectively 202677 kg, 384610 kg, 1012357 kg, 1447963 kg, and a Total inventory cost of Rp. 129231797951.
PENGEMBANGAN MOTIF FRAKTAL PADA USAHA PRODUKSI KAIN JUMPUTAN PALEMBANG Dewi, Novi Rustiana; Susanti, Eka; Hanum, Herlina; Cahyawati, Dian; Zayanti, Des Alwine
INTEGRITAS : Jurnal Pengabdian Vol 6 No 1 (2022): JANUARI - JULI
Publisher : Lembaga Penelitian dan Pengabdian kepada Masyarakat - Universitas Abdurachman Saleh Situbondo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36841/integritas.v6i1.1335

Abstract

Kain jumputan adalah salah satu kain khas Palembang. Kain jumputan Palembang memiliki ciri dan kekhasan tersendiri baik dari sisi warna maupun motif. Motif kain jumputan Palembang adalah pengulangan dari suatu bentuk bangun yang digambarkan secara berulang pada lembaran kain. Konsep pengulangan motif pada kain jumputan bersesuaian dengan konsep bangun fraktal. Penerapan teknologi komputer dan konsep matematika pada rancangan motif kain jumputan diharapkan dapat menghasilkan variasi motif kain jumputan yang dapat meningkatkan penjualan kain jumputan Palembang. Dari kegiatan pengabdian ini, dihasilkan kain dengan motif fraktal lupis dan titik tujuh. Kedua motif fraktal digambar menggunakan bangun fraktal himpunan Julia dan kurva Sierspinski Kata Kunci: Fraktal, Jumputan Palembang, Himpunan Julia, Kurva Sierspinski
AUTOMORPHISM GROUPS IN LOTUS GRAPH AND UNIFORM BOW GRAPH Khotimah, Husnul; Dewi, Novi Rustiana
Jurnal Matematika UNAND Vol. 14 No. 4 (2025)
Publisher : Departemen Matematika dan Sains Data FMIPA Universitas Andalas Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25077/jmua.14.4.311-319.2025

Abstract

This research aims to explore the connection between abstract algebraand graph theory through the study of lotus graph and uniform bow graph. The focus ison determining all automorphisms of both graphs and analyzing the algebraic structurethey form. It is shown that the set of automorphisms, under composition, satisfies thegroup axioms, thus illustrating a natural link between group theory and graph theory.Keywords: Automorphism group, lotus graph, uniform bow graph.
SOSIALISASI PENGOLAHAN SAMPAH REPLACE SEBAGAI IMPLEMENTASI MATA KULIAH PENDIDIKAN LINGKUNGAN HIDUP PADA ANAK-ANAK DI SEKITAR KAMPUS UNMUH BABEL serip, Amanda Aulia Rachman; Dian Lestari; Pipin Palinda; Lailatul Karomah; Fitri Yani; Novi Rustiana Dewi
AbdiMuh : Jurnal Pengabdian Masyarakat Vol 6 No 1 (2025): AbdiMuh
Publisher : Unmuh Babel Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35438/abdimuh.v6i1.243

Abstract

Waste problems and a lack of environmental awareness are crucial issues in the community, including those around campus. This study aims to promote the concept of "replace" (replacing single-use items with environmentally friendly alternatives) as part of the Environmental Education course for children around the UNMUH Babel campus. The research is motivated by the importance of instilling awareness and behavior in waste management from an early age in the younger generation. The community service method used is a participatory and educational approach, involving interactive lectures, audiovisual media, educational games, and hands-on practice in waste sorting. This activity targets elementary school-aged children in the campus environment. The results of the socialization show an increase in children's knowledge about the concept of "replace" and the importance of waste management. A post-activity evaluation showed that approximately 78% of participants were able to understand the concept of "replace," and 65% began to demonstrate positive behavioral changes such as bringing their own water bottles and sorting waste. However, challenges such as limited facilities and lack of family support remain obstacles. This study concludes that the socialization of the "replace" concept is effective in increasing children's environmental awareness and is a concrete implementation of the Environmental Education course.