Article Per Year (5 Year)
p-Index From 2021 - 2026
2.615
P-Index
This Author published in this journals
All Journal Jurnal Neutrino : jurnal fisika dan aplikasinya Jurnal Rekayasa Mesin SMARTICS Journal Jurnal SOLMA JOURNAL OF APPLIED INFORMATICS AND COMPUTING Dinamisia: Jurnal Pengabdian Kepada Masyarakat Jurnal Pendidikan Fisika dan Teknologi INTI Nusa Mandiri Journal of Mathematics Education and Science Inovasi Fisika Indonesia (IFI) Journal of Mechanical Engineering, Science, and Innovation Abdimas Indonesian Journal Scientific Journal of Informatics
Pelangi Eka Yuwita
Universitas Nahdlatul Ulama Sunan Giri
Religion Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Earth & Planetary Sciences Economics, Econometrics & Finance Education Electrical & Electronics Engineering Energy Engineering Industrial & Manufacturing Engineering Materials Science & Nanotechnology Mathematics Mechanical Engineering Physics Social Sciences Other

Published : 17 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 17 Documents
Search

 1   2 
Optimization of Random Forest Algorithm with SMOTE Method to Improve the Accuracy of Early Diabetes Prediction Nisa, Siti Khoirun; Barata, Mula Agung; Yuwita, Pelangi Eka
Scientific Journal of Informatics Vol. 12 No. 3: August 2025
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v12i3.22986

Abstract

Purpose: This research aims to examine the performance of the random forest algorithm in diabetes risk classification with data balancing using the Synthetic Minority Oversampling Technique (SMOTE) method to improve the representation of minority classes and increase the prediction accuracy value. Methods: The study used the Behavioral Risk Factor Surveillance System (BRFSS) dataset, obtained from Kaggle, which contains health-related survey data used to identify individuals at risk of diabetes. The Random Forest algorithm was applied to classify diabetes. To balance the data, the SMOTE method was used. The model’s performance was evaluated using 10-fold cross-validation by comparing result before and after SMOTE. Result: The results showed that the application of the SMOTE method improved the performance of the Random Forest classification model, especially in minority classes. Model performance in minority classes without SMOTE had poor evaluation metrics with precision of 49%, recall of 18%, and F1-score of 26%. After applying SMOTE, these values increased to precision of 96%, recall of 88%, and F1-score of 92%. Representing improvements of 47 percentage points in precision, 70 points in recall, and 66 points F1-score. The overall accuracy of the Random Forest model also increased from 86% to 92%, showing a 6 percentage point improvement. Novelty: This study use integrating the Random Forest algorithm with the SMOTE technique and validating the results using 10-fold cross-validation. The combination significantly improves minority class prediction performance in early diabetes detection, addressing the common limitations of previous studies in handling imbalanced datasets effectively.
ANALISIS OPTIK PEROVSKIT KERANG TEBALAN SEBAGAI BIO-MATERIAL SEL SURYA Rohmah, Roihatur; Yuwita, Pelangi Eka; Hilmi, Allif Rosyidy; Farikhah, Ainun Nurul
Inovasi Fisika Indonesia Vol. 14 No. 2 (2025): Vol 14 No 2
Publisher : Prodi Fisika FMIPA Universitas Negeri Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/ifi.v14n2.p202-208

Abstract

Abstrak Semikonduktor merupakan material yang terus dikembangkan, termasuk melalui pemanfaatan bahan alam sebagai kandidat bio-material sel surya. Penelitian ini berfokus pada pemanfaatan kalsium karbonat alami dari cangkang kerang tebalan Pantai Tuban sebagai bahan baku perovskit CaTiO₃ untuk material bio-sel surya, yang masih jarang diteliti. Proses sintesis dilakukan melalui tahapan pembersihan, pemanasan, dan pencampuran dalam reaksi padatan antara kalsium karbonat dan titanium dioksida (TiO₂). Analisis dilakukan menggunakan metode Tauc plot untuk menentukan celah energi serta difraksi sinar-X (XRD) guna mengidentifikasi struktur kristal. Hasil karakterisasi menunjukkan bahwa material yang terbentuk memiliki struktur kristal perovskit dengan fasa ortorombik, ditunjukkan oleh parameter kisi a = 5,388 Å, b = 5,432 Å, c = 7,639 Å, dan volume sel sebesar 223,567 ų. Selain itu, diperoleh nilai energi gap dengan transisi direct allowed (n = 1/2) sebesar 3,94 eV dan transisi indirect allowed (n = 2) sebesar 3,67 eV. Temuan ini menegaskan bahwa CaTiO₃ hasil sintesis dari cangkang kerang tebalan memiliki karakter semikonduktor yang potensial untuk diaplikasikan sebagai bio-material sel surya, khususnya sebagai aditif dalam lapisan penyangga maupun lapisan transpor elektron.   Abstract Semiconductors are materials that continue to be widely developed, including through the utilization of natural resources as potential bio-materials for solar cells. This study focuses on the use of natural calcium carbonate derived from Tebalan seashells collected from Tuban Beach as a precursor for CaTiO₃ perovskite, which remains rarely investigated. The synthesis process was carried out through several stages, namely cleaning, heating, and solid-state reaction between calcium carbonate and titanium dioxide (TiO₂). Characterization was performed using the Tauc plot method to determine the energy band gap and X-ray diffraction (XRD) to identify the crystal structure. The results show that the synthesized material possesses a perovskite crystal structure with an orthorhombic phase, indicated by lattice parameters a = 5.388 Å, b = 5.432 Å, c = 7.639 Å, and a unit cell volume of 223.567 ų. Furthermore, the band gap energy values were found to be 3.94 eV for the direct allowed transition (n = 1/2) and 3.67 eV for the indirect allowed transition (n = 2). These findings confirm that CaTiO₃ synthesized from Tebalan seashells exhibits semiconductor characteristics with promising potential for application as a bio-material in solar cells, particularly as an additive in buffer layers or electron transport layers.
Studi Pengaruh Variasi Media Pendingin Quenching dan Waktu Penahanan pada Proses Heat Treatment terhadap Kekerasan Baja AISI 1045 Yuwita, Pelangi Eka; Habib, Ahsanul; Faila, Rizka Nur
Jurnal Rekayasa Mesin Vol. 19 No. 1 (2024): Volume 19, Nomor 1, April 2024
Publisher : Mechanical Engineering Department - Semarang State Polytechnic

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32497/jrm.v19i1.4960

Abstract

Baja merupakan logam yang banyak digunakan sebagai bahan industry mesin, salah satunya yaitu baja AISI 1045. Baja AISI 1045 memiliki kandungan karbon sebesar 0,45% dan tergolong dalam jenis baja medium. Penerapan baja tersebut yaitu sebagai bahan dasar connecting rod, roda gigi, bearing serta poros, yang mana dalam pengopreasiannya mendapatkan gesekan serta tekanan yang menyebabkan keausan. Oleh karena itu, untuk menambah umur penggunaannya perlu dilakukan proses heat treatment sebagai upaya meningkatkan sifat mekanik dari baja. Dalam penelitian ini dilakukan proses heat treatment pada temperature 920℃ dengan variasi media pendingin (udara, oli SAE 15-50, air garam) dan variasi holding time (20 menit, 30 menit, 40 menit). Selanjutnya di lakukakan pengujian kekerasan menggunakan rockwell hardness test (HRB) untuk menyatakan nilai kekerasan. Hasil pengujian selanjutnya dibandingkan untuk mendapatkan nilai kekerasan yang optimal. Hasil rata ”“ rata kekerasan raw material yaitu sebesar 88,2 HRB. Selanjutnya nilai rara-rata kekerasan paling tinggi di dapatkan pada waktu penahanan 40 menit dengan media pendingin quenching ari garam sebesar 110,32 HRB dimana terjadi kenaikan dari raw material sebesar 25%. Sedangkan nilai kekerasan rata-rata paling rendah didapatkan pada waktu penahanan 20 menit dengan media quenching udara sebasar 88,9 HRB.
Implementation of ANN Optimization with SMOTE and Backward Elimination for PCOS Prediction Ilmiyah, Miftakhul; Barata, Mula Agung; Yuwita, Pelangi Eka
Scientific Journal of Informatics Vol. 12 No. 1: February 2025
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v12i1.22886

Abstract

by women, making it potentially fatal owing to delayed diagnosis and treatment. With the advent of current technology, machine learning and medical care may become associated with disease prediction. The purpose of the study is to predict PCOS using an Artificial Neural Network (ANN) Deep Learning algorithm combined with Synthetic Minority Oversampling Technique (SMOTE) for data balancing and backward elimination for feature selection, aiming to provide a more accurate diagnosis of PCOS with high accuracy from thoose combination. Methods: ANN algorithm structure with three hidden layers, each with a ReLU activation function of 128, 64, and 32 neurons, a Dropout layer, an output layer with a sigmoid activation function, and an Adam learning rate. Result: Using the SMOTE approach for data balance and backward elimination feature selection, the research attributes are reduced to 18. And ANN algorithm predicts PCOS disease achieve an accuracy of 92%. Novelty: This study uses an ANN algorithm model combined with the SMOTE data balancing technique and a feature selection method using backward elimination. These methods and techniques have proven to have high accuracy. The results of this study are expected to be used as a more accurate diagnosis by medical professionals in predicting PCOS disease.
Dye-Sensitized Solar Cells Characterization of Kitolod Leaves (Hippobroma longifora) Rohmah, Roihatur; Yuwita, Pelangi Eka; Pangestu, Prayoga Dewa; Vibianto, Erlangga Dwi Aprila
Jurnal Pendidikan Fisika dan Teknologi (JPFT) Vol 10 No 2 (2024): July - December
Publisher : Department of Physics Education, Universitas Mataram

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29303/jpft.v10i2.7469

Abstract

Energy from fossil fuels is decreasing day by day and also causes many environmental problems. In the current era, many environmentally friendly renewable energy sources have been developed, such as the development of DSSC (Dye Sensitized Solar Cells). DSSC is a technology that utilizes colored materials sourced from nature which are used as absorbers of sunlight energy to be converted into electrical energy. In this research, the synthesis and characterization of organic material from kitolod leaves as DSSC will be carried out. Kitolod leaves (Hippobroma longifora) are wild plants that usually grow in home gardens and rice fields. Kitolod leaves which have chlorophyll can support the absorption of sunlight if applied as a sensitizer in DSSC. The methods that will be used in this research are making TiO2 paste, dye solution preparation, electrolyte solution preparation, DSSC fabrication, testing and characterization. The tests that will be carried out are testing the TiO2 layer, dye absorption, and electric current. Meanwhile, the characterization that will be carried out is dye absorption using UV-Vis, and electric current using a potentiometer. From the results of light absorption shows that when the wavelength is 500 nm or the green light spectrum, the maximum light absorption by the chlorophyll of kitolod leaves is 1.78 (a.u.). The maximum voltage (Vm) and maximum current (Im) in daylight lamps have greater values ​​compared to warm white lamps. The highest efficiency of the solar light source is shown when the air mass is 1.5 (08.30-08.40). The time of DSSC immersion in dye also influences the resulting of Vm and Im values ​​which can be shown in that the Vm and Im values ​​of DSSC 2 (30 minutes) are greater than those of DSSC 1 (10 minutes). The results of open circuit voltage (Voc) and the short circuit current (Isc) in this research still need to be improved to produce better efficiency.
Analysis of the Characteristics of Cow Dung Waste Biopellets with Cassava Peel Starch Adhesive as an Alternative Fuel Kusuma, Indra; Sulistiawan, Agus; Yuwita, Pelangi Eka
Journal of Mechanical Engineering, Science, and Innovation Vol 5, No 2 (2025): (October)
Publisher : Mechanical Engineering Department - Institut Teknologi Adhi Tama Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31284/j.jmesi.2025.v5i2.8092

Abstract

This study aims to analyze the characteristics of biopellets made from the raw material of dried cow dung waste with cassava peel starch as an adhesive. The research method was carried out experimentally through several stages: raw material preparation, drying of cow dung under sunlight for 14 days, mixing with variations of cassava peel starch adhesive of 10%, 30%, and 50%, molding using a hydraulic press with pressures of 50, 60, and 70 kg, and testing the characteristics of the biopellets. The parameters tested included calorific value, moisture content, combustion rate, and ash content. The results showed that variations in adhesive levels and molding pressure affected the quality of the biopellets. The calorific value obtained ranged from 2,795.4 to 3,595.9 cal/g, moisture content 9.5–12.5%, combustion rate 0.74–1.71 g/min, and ash content 0.90–1.20%. In conclusion, biopellets from cow dung with the addition of cassava peel starch adhesive have characteristics that are feasible to be developed as an environmentally friendly alternative fuel.
Pelatihan dokumen Digital Land Document Application Training Towards a Digital Village at the Kujung Village Government : Pelatihan Aplikasi Dokumen Pertanahan Menuju Desa Digital Pada Pemerintahan Desa Kujung Nurma*, Nur Mahmudah; Ningrum, Ifa Khoiria; Yuwita, Pelangi Eka
Dinamisia : Jurnal Pengabdian Kepada Masyarakat Vol. 8 No. 6 (2024): Dinamisia: Jurnal Pengabdian Kepada Masyarakat
Publisher : Universitas Lancang Kuning

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31849/dinamisia.v8i6.23313

Abstract

Digital-based data management is essential for village development, particularly in managing digital land documents. These documents significantly enhance the efficiency of data management, making it easier for residents to search, input, and update information. The primary objective of our community service initiative is to equip Kujung Village government employees with the necessary knowledge and skills for effective digital land document management. Our implementation approach involves three stages: execution, evaluation, and monitoring to ensure program sustainability. On September 26, 2024, at Kujung Village Hall, 15 village officials, 4 BPD members, and 6 LKMD members participated enthusiastically in the program. The results indicated a strong consensus among participants regarding the proposed indicators, and a notable increase in their understanding. Consequently, village officials showed a greater inclination to utilize digital applications for data management over traditional manual methods. This initiative has also led to improved services that are faster, more transparent, and accountable
Co-Authors Aditya, Ziedan Agus Sulistiawan Amalia, Salsabila Dani Dina, Intan Rachma Dwi Tiyas Novitasari Faila, Rizka Nur Farikhah, Ainun Nurul Habib, Ahsanul Hilmi, Allif Rosyidy Ifa Khoiria Ningrum Ilmiyah, Miftakhul Indra Kusuma Maulani, Vicka Rizqi Midya Yuli Amreta Mula Agung Barata Nisa, Siti Khoirun Nur Mahmudah Nur Mahmudah Nurma*, Nur Mahmudah Pangestu, Prayoga Dewa Roihatur Rohmah Sa’diyah, Zumrotus Togik Hidayat Vibianto, Erlangga Dwi Aprila