DATA PREPROCESSING AND CLASSIFICATION ALGORITHMS USING ENSEMBLE METHODS BASED ON DECISION TREES

A. Akhatov, F. Nazarov

Samarqand davlat universiteti ilmiy tadqiqotlar axborotnomasi · 2026-yil

Annotatsiya

This research paper proposes a methodology for processing datasets based on modern methods. Iterative imputation based on decision trees was used to systematically replace missing values in the dataset. The remaining irregular part of the missing values was filled in using the SVD imputation method based on Matrix Factorization. The data's extreme values and outliers were stabilized using a logarithmic transformation. The prepared data was trained on decision trees and ensemble models based on them. The effectiveness of the proposed ensemble approaches in classifying crop types based on soil data was evaluated on training, validation, and test sets, and the results were analyzed.

Maqola ma’lumotlari
MualliflarA. Akhatov, F. Nazarov
JurnalSamarqand davlat universiteti ilmiy tadqiqotlar axborotnomasi
Nashr sanasi2026-01-01
Jild1
Son1
Betlar65-74
DOI10.59251/2181-3973.2025.v1.138.1.3759

Samarqand davlat universiteti ilmiy tadqiqotlar axborotnomasi jurnalidan boshqa maqolalar

Samarqand davlat universiteti ilmiy tadqiqotlar axborotnomasi — barcha maqolalar