<![CDATA[This study aims to analyze and compare the performance of Gaussian Naïve Bayes and K-Nearest Neighbor (KNN) algorithms in classifying the feasibility of construction materials to support decision-making in construction projects. A quantitative comparative study design was applied using 127 samples of structural building materials collected from 15 contractor companies in Lhokseumawe City, Indonesia. The dataset consists of five predictor variables: price, compressive strength, water absorption, delivery time, and supplier rating. Data preprocessing included missing value imputation, outlier handling using the interquartile range method, normalization using Min-Max scaling, and class balancing using Synthetic Minority Over-sampling Technique (SMOTE). Model evaluation was conducted using accuracy, precision, recall, F1-score, and AUC, while feature importance was analyzed using permutation importance. The results show that the KNN model (k = 5) outperforms Gaussian Naïve Bayes across all evaluation metrics, achieving an accuracy of 92.11% and an AUC of 0.934.]]>
Copyrights © 2026
