An Innovative Method to Predict Real Estate Prices Using Convolutional Block Attention Module

Both real estate brokers and owners can benefit strategically from foresight into rising or falling property values. This study intends to develop a novel method to accurately forecast real estate prices using machine learning models and actual transaction data. In machine learning models, the attributes of the real estate and the transaction prices are the independent variables and the dependent variables, respectively. Preprocessing, feature selection, and model training are used in the proposed method. Preprocessing involves the use of normalization. The selection of features is based on PCA and sequential analysis. CBAM is the model that is utilized for training. When compared to GRU and LSTM, two additional models, the proposed technique performs well.