Data-Driven Insights in Real Estate: Property Recommendation System and House Rent Prediction using Machine Learning

In the fast-changing market of real estate, making an accurate prediction of house rent, along with proper recommendations of properties based on user preferences, is crucial for buyers and agents. In this research, we compared the performance of four models: Linear Regression, Random Forest Regression, SVM, and Decision Tree Regressor, to predict house rent. In addition, we have applied the KNN algorithm as well as the Cosine similarity measure to make property recommendations by using user preferences. Also, we tested the performance of each predictive model by computing Mean Squared Error and R-Squared. The effectiveness of the recommendation system is tested in terms of its property suggestions matching to the preferred properties of users. The results show the outputs for each of the machine learning algorithms and how a combination of predictive model and recommendation techniques is used to improve the user experience in real estate.