Prediction of Boarding House Rental Prices Using Multiple Linear Regression Method

Given the significant influx of students to Surabaya, there is a high demand for affordable boarding house rental near universities. One of the issues that arises is figuring out the rent fee set by the landlord.. This study will focus on solving the problem using machine learning with Multiple Linear Regression (MLR) method. This study also focuses on developing a predictive model for temporary housing rental prices around the National Development University "Veteran" Java. Key variables include rental price, room type, room size, availability of air conditioning, WiFi, private bathrooms, kitchen access, 24-hour access, and distance to the university. The dataset was split into training and testing sets (80:20 ratio) for model development and evaluation. The MLR model achieved an R² value of 0.76, an RMSE of 211555.8, and a MAPE of 0.18, indicating high predictive accuracy.