Article Per Year (5 Year)
p-Index From 2020 - 2025
2.494
P-Index
This Author published in this journals
All Journal E-Jurnal Matematika JURNAL MATEMATIKA STATISTIKA DAN KOMPUTASI Jurnal Sains Matematika dan Statistika Format : Jurnal Imiah Teknik Informatika EIGEN MATHEMATICS JOURNAL International Journal of Engineering, Technology and Natural Sciences (IJETS) Abdimasku : Jurnal Pengabdian Masyarakat Jurnal Inovasi Penelitian dan Pengabdian Masyarakat Jurnal Ilmu Komputer dan Informatika Jurnal Eurekamatika UPGRADE: Jurnal Pendidikan Teknologi Informasi Diophantine Journal of Mathematics and Its Applications
Moch Anjas Aprihartha
Unknown Affiliation
Civil Engineering, Building, Construction & Architecture Computer Science & IT Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Health Professions Industrial & Manufacturing Engineering Mathematics Social Sciences Other

Published : 18 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 18 Documents
Search

 1   2 
Optimization of Classification Algorithms Performance with k-Fold Cross Validation Aprihartha, Moch. Anjas; Idham, Idham
Eigen Mathematics Journal Vol 7 No 2 (2024): December
Publisher : University of Mataram

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29303/emj.v7i2.212

Abstract

Supervised learning is a predictive method used to make predictions or classifications. Supervised learning algorithms work by building a model using training data that includes both independent and dependent variables. Several methods for building classification include Logistic Regression, Naive Bayes, K-Nearest Neighbor (KNN), decision tree, etc. The lack of capacity of a classification algorithm to generalize certain data can be associated with the problem of overfitting or underfitting. K-fold cross-validation is a method that can help avoid overfitting or underfitting and produce a algorithm with good performance on new data. This study will test the Naive Bayes, K-Nearest Neighbor (KNN), Classification and Regression Tree (CART), and Logistic Regression methods with k-fold cross-validation on two different datasets. The values of k set for cross-validation are 2, 3, 5, 7, and 10. The analysis results concluded that each classification algorithm performed best at 10-fold cross-validation. In DATA 1, the Naive Bayes algorithm has the highest average accuracy of 0.67 (67%) and the error rate is 0.33 (33%), followed by the CART algorithm, KNN, and finally logistic regression. While DATA 2, the KNN algorithm has the highest average accuracy of 0.66 (66%) and an error rate of 0.34 (34%), followed by the CART algorithm, Naive Bayes, and finally logistic regressionbut can be a reference if you want to predict the growth direction of the accommodation and food service activities sector.
Comparison of Discrete Adaptive Boosting Algorithms for Classification and Regression Tree and Naive Bayes in Pistachio Nut Classification Aprihartha, Moch. Anjas; Azzahro, Salwa Paramita; Azizah, Rahmatul; Andrianza, Muhammad Rafly
Jurnal Internasional Teknik, Teknologi dan Ilmu Pengetahuan Alam Vol 7 No 1 (2025): International Journal of Engineering, Technology and Natural Sciences
Publisher : Universitas Teknologi Yogyakarta, Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46923/ijets.v7i1.396

Abstract

Machine learning is an effective tool for identifying and classifying various conditions, such as predicting shoe sales, classifying raisin types, classifying fruit productivity, and so on. This technique is widely used in various sectors. One example is pistachio sorting. In some places, pistachio sorting is still done traditionally by humans. This is disadvantageous because the costs tend to be high, and the sorting process becomes inconsistent and less effective. The use of machine learning algorithms can be a breakthrough in overcoming this problem. Naive Bayes and Classification and Regression Tree (CART) are machine learning algorithms commonly used in the classification process. To improve classification accuracy, these two basic models are integrated with the Discrete Adaptive Boosting (Discrete AdaBoost) algorithm. This study aims to assess the effectiveness of machine learning algorithms in identifying the characteristics of pistachios. Algorithm testing was carried out using the k-fold cross-validation technique. The estimated average performance results of all classification models do not show significant differences. The Discrete AdaBoost CART model has the best accuracy, specificity, and f1-score, at 86.49%, 85.78%, and 88.32%, respectively. Therefore, the Discrete AdaBoost CART model is a suitable model for classifying pistachio types. This shows that ensemble approaches such as Discrete AdaBoost CART can make a significant contribution to improving the performance of classification systems, especially in the context of data with many relevant features. This study was limited to identifying binary classes of pistachios. In further research, it is recommended to explore machine learning algorithms for multiclass of pistachio nuts.
Perbandingan Algoritma Real Adaptive Boosting pada Regresi Logistik, CART, dan Naive Bayes dalam Klasifikasi Biji Labu Aprihartha, Moch Anjas; Fallo, Sefri Imanuel; Rasikhun, Hady
Jurnal Sains Matematika dan Statistika Vol 11, No 2 (2025): JSMS Juli 2025
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/jsms.v11i2.36859

Abstract

Labu merupakan spesies tanaman yang bernilai ekonomis dan medis. Hampir setiap bagian dari labu dapat dikonsumsi terutama pada bijinya. Minyak dari biji labu dapat juga digunakan sebagai saus untuk salad, produk kosmetik, sabun dan lilin. Keterampilan dalam mengklasifikasikan biji labu dengan tepat sangat dibutuhkan diberbagai sektor, seperti pertanian dan industri pangan. Dibutuhkan teknologi pengembangan yang dapat mengidentifikasi dan mensortir biji labu dengan mudah dan cepat. Beberapa algoritma yang umum dapat digunakan untuk mengidentifikasi jenis biji labu seperti algoritma regresi logistik (RL), Classification and Regression Tree (CART), dan Naive Bayes (NB). Penelitian ini bertujuan mengeksplorasi model RL, CART, dan NB pada dua jenis varietas biji labu, yaitu Ürgüp Sivrisi dan Çerçevelik berdasarkan karakteristik fisiknya. Selain itu, digunakan pendekatan Real Adaptive Boosting (RAB) untuk meningkatkan kinerja model dasar. Teknik ini bekerja dengan kemampuan menggabungkan beberapa model homogen secara berulang untuk menghasilkan model yang kuat. Hasil uji kinerja model klasifikasi diperhitungkan melalui metrik evaluasi. Model RAB-RL memiliki performa tertinggi pada akurasi, presisi, dan f1-score sehingga menjadikan model terbaik dalam mengklasifikasikan jenis biji labu dibandingkan model-model lainnya. Dalam model dasar, model RL memiliki performa terbaik dibawah model RAB-RL
Klasifikasi Produktivitas Buah Nanas Menggunakan Algoritma Classification and Regression Tree (CART) Aprihartha, Moch. Anjas; Putrawan, Zulhandi; Zulhan , Dicky; Ahardika Nurfaizal, Fatma
Diophantine Journal of Mathematics and Its Applications Vol. 3 No. 1 (2024)
Publisher : UNIB Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33369/diophantine.v3i1.34193

Abstract

Indonesia is one of the countries that has a variety of fruits cultivated. One of them is the pineapple fruit. Various pineapple-based products such as pineapple juice, canned foods, pineapple jam, etc. The high demand for pineapples presents an opportunity for companies to increase pineapple product processing. The increase in pineapple productivity is influenced by several factors, one of which is the extent of land and the type of pineapple produced. To improve pineapple productivity, it can be done by classifying the types of pineapples based on productive and non-productive categories. The purpose of this classification is to enable farmers or plantation managers to allocate resources more efficiently by providing more intensive care for productive category pineapples. The classification method that can be used to classify productive and non-productive pineapples is the Classification and Regression Tree (CART) algorithm. The CART method is a method that produces decision tree models that are used to solve classification and regression problems. This research uses the CART method to classify pineapple productivity. The research results obtained accuracies, sensitivities, specificities, and precisions of 97.06%; 92.31%; 100%; 100% respectively. Meanwhile, the AUC obtained is 0.962 which indicates that the model is very good at predicting pineapple productivity correctly.
Eksplorasi Penggunaan Platform Kahoot! Sebagai Media Belajar Matematika Pada Siswa Kelas VIII SMP Aprihartha, Moch. Anjas; Prabowo, Wahyu Aji Eko; Wibawa, Vincentius Hadianta
Jurnal Inovasi Penelitian dan Pengabdian Masyarakat Vol. 4 No. 2 (2024): Desember
Publisher : Indonesia Emerging Literacy Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.53621/jippmas.v4i2.383

Abstract

Pembelajaran merupakan mekanisme penyerapan ilmu pengetahuan yang terjadi di kelas. Dalam prosesnya terdapat aktivitas siswa dan guru yang didukung oleh media, alat, metode, serta bahan ajar yang sesuai dengan kebutuhan di kelas. Permasalahan yang dialami dalam proses belajar pada umumnya dapat berdampak signifikan pada efektifitas pendidikan dan pencapaian hasil belajar. Oleh karena itu diperlukan adopsi teknik belajar yang lebih dinamis dan interaktif agar proses belajar menjadi menyenangkan. Salah satunya dengan memanfaatkan platform Kahoot!. Platform Kahoot! merupakan website yang menyediakan media pembelajaran berbasis permainan interaktif. Pengabdian ini bertujuan untuk mengevaluasi kesesuaian penggunaan platform Kahoot! dalam pengembangan metode belajar serta akan dilakukan analisis sejauh mana respon siswa terhadap penggunaan fitur interaktif platform ini. Pelaksanaan kegiatan pengabdian ini berlokasi di SMPN 3 Gunung Sari dengan mengambil sampel siswa kelas VIII. Metode yang diterapkan meliputi studi lapangan, melatih siswa menggunakan Kahoot!, dan evaluasi hasil belajar. Secara keseluruhan dari hasil evaluasi yang telah dilaksanakan, siswa kelas VIII SMPN 3 Gunung Sari sepakat bahwa platform Kahoot! dapat memberikan pengalaman baru yang menyenangkan dalam menjalankan proses belajar matematika, terutama dari segi tampilan dan kejelasan soal yang dapat dipahami oleh peserta didik.
Algoritma Synthetic Minority Oversampling Technique dan C5.0 dalam Mengatasi Ketidakseimbangan Data pada Klasifikasi Kelulusan Siswa Aprihartha, Moch Anjas; Putrawan, Zulhandi; Zulhan, Dicky; Nurfaizal, Fatma Ahardika
Upgrade : Jurnal Pendidikan Teknologi Informasi Vol 2 No 1 (2024): Vol. 2 No. 1 Agustus 2024
Publisher : Pendidikan Teknologi Informasi Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/upgrade.v2i1.4148

Abstract

Algoritma supervised learning digunakan untuk memprediksi dan mengklasifikasikan atribut tertentu, namun masalah utama adalah distribusi data yang tidak merata antar kelas yang dapat menyebabkan overfitting. Untuk mengatasi ini, diperlukan augmentasi kelas minoritas menggunakan teknik Synthetic Minority Oversampling Technique (SMOTE). Tujuan penelitian ini memberikan solusi praktis untuk mengatasi ketidakseimbangan data dengan SMOTE pada kasus siswa yang tidak lulus semua mata pelajaran, guna mengurangi risiko overfitting. Metode penelitian ini adalah penelitian eksperimental dengan pendekatan kuantitatif menggunakan data sekunder dari kelulusan mata pelajaran siswa. Teknik analisis data hasil SMOTE diuji dengan algoritma C5.0, dan variasi state 1 hingga 100 digunakan untuk memastikan pemilihan data training dan testing secara acak di setiap iterasi. Hasil penelitian menunjukkan bahwa uji data asli dengan algoritma C5.0 menghasilkan plot akurasi, recall, dan spesifisitas yang tidak konsisten, sedangkan uji data yang diolah dengan SMOTE menunjukkan plot yang stabil mendekati 100%. Artinya, data SMOTE memberikan performa yang lebih baik pada algoritma C5.0 dibandingkan data asli. Efektivitas teknik SMOTE dan algoritma C5.0 dapat berkontribusi bagi peneliti yang menghadapi masalah serupa. Implikasi hasil penelitian ini juga dapat dijadikan acuan dalam membuat aplikasi untuk mendeteksi kelulusan siswa guna mempermudah guru dalam mengambil keputusan.
Algoritma Klasifikasi Naive Bayes dalam Identifikasi Pasien Demam Berdarah Dangue (DBD) Aprihartha, Moch Anjas; Aprihartha, Moch. Anjas; Putrawan, Zulhandi; Zulhan, Dicky
Upgrade : Jurnal Pendidikan Teknologi Informasi Vol 3 No 1 (2025)
Publisher : Pendidikan Teknologi Informasi Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/upgrade.v3i1.5457

Abstract

Dengue fever (DF) is an illness caused by the Dengue virus, transmitted to humans through the bite of female Aedes aegypti mosquitoes, and the rise in DF cases often leads to a surge in hospital visits that can result in shortages of beds and medical personnel. In severe conditions, patients require treatment from health professionals experienced in managing this disease, and with advancements in scientific methods, classification techniques have become essential in identifying the severity level of DF patients to determine immediate and necessary treatment. This study aims to classify DF patients who require inpatient care by applying the Naive Bayes method to 230 observation records obtained from medical data of DF patients at Anwar Makkatutu Hospital in Bantaeng Regency during the 2019–2020 period, with model performance evaluated using a confusion matrix. The findings show that the Naive Bayes algorithm demonstrates fairly good performance in identifying patients who need hospitalization and those who do not, indicated by its AUC, accuracy, sensitivity, and specificity values of 0.702, 70.11%, 59.09%, and 81.40%, respectively. These results support more efficient allocation of limited healthcare resources and offer practical implications for clustering DF patients who require medical attention, enabling health authorities to improve planning, prepare adequate medical facilities, and optimize treatment readiness, while also contributing valuable insights to the scientific literature on related topics.
Algoritma Klasifikasi Naive Bayes dalam Identifikasi Pasien Demam Berdarah Dangue (DBD) Aprihartha, Moch Anjas; Aprihartha, Moch. Anjas; Putrawan, Zulhandi; Zulhan, Dicky
Upgrade : Jurnal Pendidikan Teknologi Informasi Vol 3 No 1 (2025)
Publisher : Pendidikan Teknologi Informasi Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/upgrade.v3i1.5457

Abstract

Dengue fever (DF) is an illness caused by the Dengue virus, transmitted to humans through the bite of female Aedes aegypti mosquitoes, and the rise in DF cases often leads to a surge in hospital visits that can result in shortages of beds and medical personnel. In severe conditions, patients require treatment from health professionals experienced in managing this disease, and with advancements in scientific methods, classification techniques have become essential in identifying the severity level of DF patients to determine immediate and necessary treatment. This study aims to classify DF patients who require inpatient care by applying the Naive Bayes method to 230 observation records obtained from medical data of DF patients at Anwar Makkatutu Hospital in Bantaeng Regency during the 2019–2020 period, with model performance evaluated using a confusion matrix. The findings show that the Naive Bayes algorithm demonstrates fairly good performance in identifying patients who need hospitalization and those who do not, indicated by its AUC, accuracy, sensitivity, and specificity values of 0.702, 70.11%, 59.09%, and 81.40%, respectively. These results support more efficient allocation of limited healthcare resources and offer practical implications for clustering DF patients who require medical attention, enabling health authorities to improve planning, prepare adequate medical facilities, and optimize treatment readiness, while also contributing valuable insights to the scientific literature on related topics.
Co-Authors Achmad Wahid Kurniawan Ahardika Nurfaizal, Fatma Akrom, Muhamad Alam, Taufik Nur Andrianza, Muhammad Rafly Ayu Pertiwi Azizah, Rahmatul Azzahro, Salwa Paramita Budi, Setyo Dicky Zulhan Fatma Ahardika Nurfaizal Fitri Astutik Hady Rasikhun Harun Al Azies Herowati, Wise Hidayat, Novianto Nur Husniyadi, M. Idham Idham Ignatius Moses Setiadi, De Rosal JUS PRASETYA Jus Prasetya Muhammad Naufal, Muhammad Muljono Muljono Nani Sulistianingsih Noor Ageng Setiyanto, Noor Ageng Nurfaizal, Fatma Ahardika Prabowo, Wahyu Aji Eko Prasetya, Jus Putrawan, Zulhandi Safitri, Aprilyani Nur Sefri Imanuel Fallo Sudibyo, Usman Supriadi Rustad T. Sutojo Trisnapradika, Gustina Alfa Wibawa, Vincentius Hadianta Zulhan , Dicky Zulhan, Dicky Zulhandi Putrawan