Article Per Year (5 Year)
p-Index From 2021 - 2026
0.817
P-Index
This Author published in this journals
All Journal Jurnal Nasional Komputasi dan Teknologi Informasi Jambura Journal of Electrical and Electronics Engineering
Rahmat Thaib
Universitas Pohuwato
Aerospace Engineering Automotive Engineering Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Energy Engineering Neuroscience Transportation

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

 1 
Klasifikasi Pasien Persalinan Caesar Menggunakan Metode Nave Bayes Berbasis Forward Selection Muh Faisal; Bahrin Dahlan; Rahmat Thaib
Jurnal Nasional Komputasi dan Teknologi Informasi (JNKTI) Vol 6, No 6 (2023): Desember 2023
Publisher : Program Studi Teknik Komputer, Fakultas Teknik. Universitas Serambi Mekkah

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32672/jnkti.v6i6.7143

Abstract

Abstrak - Hasil observasi di lingkungan rumah maupun kantor yakni cukup banyak ibu hamil yang akhirnya melakukan operasi caesar. Ada beberapa penyebab seorang ibu hamil melakukan caesar. Pertama, faktor kesehatan ibu. Kedua, faktor janin. Ketiga adalah faktor gabungan dari faktor ibu dan janin. Faktor-faktor tersebut menjadi indikasi apakah persalinan akan dilakukan dengan mutlak atau mungkin juga bisa menjadi relatif. Pada penelitian ini akan menerapkan metode Naive Bayes dengan optimasi forward selection untuk mendapatkan klasifikasi persalinan caesar dengan lebih optimal dimana hasil yang didapatkan penelitian sebelumnya terhadap prediksi ibu melahirkan hanya mendapat akurasi 88%. Nave Bayes merupakan pengklasifikasian dengan metode probabilitas dan statistik. Dari sembilan atribut yang digunakan yaitu Gravid Aterm, Riwayat SC, Posisi Bayi, Bayi Besar, Plasenta, Ketuban, Penyakit Ibu, Gemelli, dan Inpatu lalu dengan menggunakan algoritma Naive Bayes berbasis Forward Elimination didapatkan empat atribut weight yaitu Gravid Aterm, Bayi Besar, Ketuban dan t Gemelli dalam mengklasifikasi partus atau persalinan caesar. Secara mandiri tingkat akurasi yang dihasilkan algoritma Naive Bayes adalah 93,33 %. Sedangkan dengan menambahkan seleksi fitur Forward Elimination menghasilkan akurasi 94% dalam klasifikasi pasien persalinan Caesar. Dengan demikian, metode Naive Bayes berbasis Forward Elimination dapat digunakan sebagai metode yang lebih optimal dari penelitian sebelumnya.Kata Kunci: Nave Bayes, Forward Selection, CaesarAbstract -The results of observations in the home and office environment are quite a lot of pregnant women who end up doing cesarean sections. There are several causes of a pregnant woman doing a cesarean. First, the maternal health factor. Secondly, fetal factors. Third is the combined factor of maternal and fetal factors. These factors are an indication of whether labor will be done absolutely or maybe it can also be relative. This study will apply the Naive Bayes method with forward selection optimization to get a more optimal classification of cesarean delivery where the results obtained by previous studies on the prediction of childbirth only got 88% accuracy. Nave Bayes is a classification by probability and statistical methods. Of the nine attributes used, namely Gravid Aterm, SC History, The position of the Baby, Big Baby, Placenta, Amniotic, Maternal Disease, Gemelli, and Inpatu then using the Naive Bayes algorithm based on Forward Elimination obtained four weight attributes, namely Gravid Aterm, Big Baby, Amniotic and Gemelli t in classifying partus or cesarean delivery. Meanwhile, by adding the Forward Elimination feature selection resulted in 94% accuracy in the classification of Cesarean delivery patients. Thus, the Naive Bayes method based on Forward Elimination can be used as a more optimal method than previous studies.Keywords : Nave Bayes Forward Selection Caesar
Pengelompokan Tingkat Keaktifan Siswa dalam Mengikuti Proses Belajar di SMA Negeri 2 Tilamuta Menggunakan Metode K-Means Clustering Muh Faisal; Hamsir Saleh; Rahmat Thaib; Ismail Tolotu
Jurnal Nasional Komputasi dan Teknologi Informasi (JNKTI) Vol 8, No 5 (2025): Oktober 2025
Publisher : Program Studi Teknik Komputer, Fakultas Teknik. Universitas Serambi Mekkah

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32672/jnkti.v8i5.9829

Abstract

Abstrak - K-Means clustering adalaha salah satu metode pengelompokan data yang sering digunakan dalam analisis data berbasis pembelajaran mesin (machine learning). Metode ini bertujuan untuk membagi data kedalam beberapa cluster berdasarkan kemiripan antar data dalam ruang multidimensi. Prose k-means dimulai dengan menentukan jumlah cluster yang diinginkan, lalu algoritma mengelompokan data secra iteratif kedalam cluster berdasarkan jarak terdekat ke titik pusat cluster (centroid), hingga posisi centroid stabil dan ada perubahan signifikan. Dalam konteks pendidikan, metode ini digunakan untuk mengelompokan siswa berdasarkan tingkat keaktifan mereka di dalam kelas atau pembelajaran. Data yang digunakan meliputi berbagai indikator keaktifan, seperti partisipasi dalam pembelajaran (diskusi dan tanya jawab), keterlibatan dalam pengerjaan tugas kelompok, pemahaman terhadap materi yang diberikan, ketertarikan pada guru bidang studi, dan juga ketertarikan pada mata pelajaran. Dengan k-means clustering, siswa dapat dikelompokan kedalam 3 kategori seperti kelompok sangat aktif, aktif dan tidak aktif. Berdasarkan hasil analisis menunjukan bahwa sebagian vesarsiswa berada dalam cluster tidak aktif, sementara beberapa siswa lainnya tersebar di cluster aktif, dan sangat aktif. Pengelompokan ini membantu guru dalam memahami variasi tingkat partisipasi siswa dan merancang strategi pembelajaran yang lebih efektif untuk setiap kelompok. Dengan demikian, metode k-means clustering tidak hanya berguna untuk analisis data, tetapi juga dapat diimplementasikan sebagai alat pendukung dalam pembelajaran.Kata Kunci: K-Means Clustering; pengelompokan siswa; tingkat keaktifan; Pembelajaran di kelas; analisis data dan pendidikan; Abstract - K-means clustering is a popular data clustering method used in statistical analysis and machine learning. This method aims to partition data into several clusters based on similarities in a multidimensional space. The k-means process starts by determining the desired number of clusters (k), after which the algorithm iteratively groups data points into clusters based on their proximity to the nearest cluster centroid, continuing until the centroids stabilize and no significant changes occur. In the context of education, this method can be applied to group students based on their level of classroom engagement. The data used includes various indicators of engagement, such as participation in discussions, the number of questions asked, involvement in group tasks, and attendance rates. Using k-means clustering, students are categorized into several groups, including highly active, active, moderately active, and less active. The analysis results show that the majority of students fall into the moderately active cluster, while a smaller number of students are distributed across the highly active and less active clusters. This clustering helps teachers better understand the variations in student participation and design more effective instructional strategies tailored to each group. Thus, k-means clustering is not only useful for data analysis but can also be implemented as a supportive tool in crafting more targeted educational planning.Keywords: k-means clustering; student clustering; engagement levels; classroom learning; educational data analysis;
Classification of Pneumonia Severity in Children Using the Fuzzy K-Nearest Neighbor Method Based on Patient Clinical Data Rahmat Thaib; Betrisandi Betrisandi
Jambura Journal of Electrical and Electronics Engineering Vol 8, No 2 (2026): Juli - Desember 2026
Publisher : Electrical Engineering Department Faculty of Engineering State University of Gorontalo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37905/jjeee.v8i2.34624

Abstract

Pneumonia is one of the most deadly acute respiratory infections in children, especially in the toddler age group. Indonesia ranks eighth among 15 countries with the highest pneumonia mortality rate, namely 22,000 toddler deaths per year. Pneumonia can be caused by various microorganisms such as viruses, fungi, and bacteria. The occurrence of pneumonia is characterized by symptoms of cough and/or difficulty breathing such as rapid breathing and lower chest wall indrawing. The diagnosis of pneumonia is generally based on a combination of clinical symptoms such as fever, cough, rapid breathing, and physical examination results such as physical or radiological, however, the diagnostic process often encounters obstacles, such as limited trained medical personnel, limited diagnostic tools and subjectivity in assessing symptoms, especially in children who are not yet able to communicate their complaints clearly. This study aims to classify pneumonia based on symptoms and severity, namely severe pneumonia and mild pneumonia in children to assist medical personnel in making more accurate and efficient decisions. The results of this study indicate that the Fuzzy K-Nearest Neighbor method with k=3 and m=2 produces an accuracy of 62.67%, precision of 65.91%, recall of 69.05%, F1-Score of 67.44%, and a deviation of ±8.00% in classifying pneumonia in children.
Comparison of NBC and KNN in Classifying Stunting in Children in Rural Areas Betrisandi Betrisandi; Rahmat Thaib
Jambura Journal of Electrical and Electronics Engineering Vol 8, No 1 (2026): Januari - Juni 2026
Publisher : Electrical Engineering Department Faculty of Engineering State University of Gorontalo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37905/jjeee.v8i1.34488

Abstract

Stunting is one of the chronic nutritional problems that remains a serious concern in Indonesia. Children who experience stunting not only experience physical growth retardation, but also cognitive development disorders that have the potential to reduce intelligence, academic achievement, and productivity in adulthood. The problem in this study is the high prevalence of stunting in children in rural areas. The purpose of this study is to analyse the performance of the Naïve Bayes Classifier (NBC) and K-Nearest Neighbour (KNN) and compare the performance of the two methods to determine the most optimal method for classifying stunting status in children in accordance with the Research Master Plan with a focus on engineering and technology for improving ICT content and the research topic of big data technology development. The research methods used included data collection through observation and interviews. Data processing and analysis were carried out by comparing the NBC and KNN methods in classifying child stunting. The results of this study indicate that the NBC method has higher accuracy, namely 95.24% and an F1-score of 97%, compared to the KNN method, which has an accuracy of 76.19% and an F1-score of 86%. Therefore, the KNN method is more optimal for use in classifying stunting in children.
Co-Authors Bahrin Bahrin Betrisandi, Betrisandi Hamsir Saleh Ismail Tolotu Muh Faisal Muh Faisal