<![CDATA[The increasing use of computer science systems in various fields of work is driving the growth of laptop. There are many choices of laptop products with different specifications. Laptop price predictions can help consumers to find out the price range of laptops according to the laptop features they want. This study aims to apply the random forest regression method to predict laptop price based on laptop features. The data is splitted into 2 parts, 1130 data as training set and 283 data as testing set. Hyperparameters tuning was performed using random search CV with 10 fold cross-validation to find the combination of hyperparameters that produced the optimal model. The best hyperparameters obtained to build a random forest regression model are ntree = 400, mtry = 2, and nodesize = 2. The MAPE value of the testing set for the random forest regression model is 14,3% which indicates the performance of the model has good forecasting ability. The results of the analysis show that the random forest regression method can be applied to predict laptop prices based on laptop features. Based on the variable importance, the variable that has the greatest contribution to the laptop price prediction results is RAM with VI = 0,359.]]>
Copyrights © 2025
