Klasifikasi Sentimen Menggunakan Algoritma K-NN Berdasarkan Pendekatan Random Search Cross Validation

Abstract

This research discusses sentiment classification using the K-Nearest Neighbor (K-NN) algorithm with the RANDOM SEARCH CROSS VALIDATION (RSCV) approach. This method aims to classify ChatGPT app user reviews into three categories: positive, negative, and neutral. In this study, review data is collected from the Play Store and text preprocessing is performed. Furthermore, the K-NN model will use RSCV in determining parameter values. The evaluation results show that the K-NN model with the best parameters can classify the sentiment of reviews with good accuracy.