Comparison of Random Forest and Naïve Bayes Methods for Classifying and Forecasting Soil Texture in the Area Around Das Kalikonto, East Java

Abstract

Soil texture is used to determine airflow, heat, instability, water holding capacity, and the shape and structure of the soil structure. Soil texture as an important attribute that determines the direction of soil management must be modeled accurately. However, soil texture is a soil attribute that is quite difficult to model. It is a compositional data set that describes the particle size of the soil mineral fraction (sand, silt, and clay). The methods used to classification and predict soil texture with machine learning algorithms are Random Forest (RF) and Naïve Bayes (NB). The purpose of this study was to classify the distribution of soil texture using the Random Forest and Naïve Bayes methods to obtain the most accurate grouping results. This research was conducted in the area around Kalikonto River Basin, East Java Province. The performance-based tests show that the RF algorithm provides higher accuracy in predicting soil texture based on the Digital Elevation Model (DEM). The results of RF’s performance testing on training data and testing data gave an accuracy value of 92.55% and 87.5%. Classification using the Naïve Bayes method produces an accuracy value of 89.98% on testing data and 80.65% accuracy on training data.

Designing A Pdf Malware Detection System Using Machine Learning

“…Random Forest's dependability is applicable in a variety of settings, such as the Internet of Things (IoT) [22]. Previous research [23], [24], [25], [26], [27], has shown that the Random Forest algorithm is reliable when it comes to classification; nevertheless, this study specifically highlights the Random Forest algorithm's use in the context of harmful PDF classification.…”

Komparasi Akurasi Metode Random Forest Dan Deep Learning pada Reservasi Hotel

“…Menurut (Pramoedyo, 2022) Random Forest adalah pengembangan pohon keputusan dengan menggunakan banyak pohon keputusan dimana setiap pohon keputusan telah dilatih menggunakan sampel individu dan setiap atribut dibagi menjadi pohon yang elit antara satu set atribut acak dan setiap atribut dibagi menjadi pohon yang dipilih antara subset atribut acak dan setiap atribut dibagi menjadi pohon yang dipilih antara satu set atribut acak dan setiap atribut Random Forest mungkin merupakan teknik organisasi yang mengumpulkan variabel freelance masih sebagai data sampel, yang mengarah ke pohon klasifikasi dengan berbagai ukuran dan bentuk.…”