This document discusses predicting house prices in Tehran, Iran using housing data. It explores factors that impact price like size, area, location, bedrooms and bathrooms. The dataset has 3479 records and 8 variables, including categorical variables like parking and continuous variables like area and price. The proposed solution is to process the dataset, build a regression model, train it and test it to make predictions. Implementation steps include data cleaning, feature engineering, outlier removal, model building, testing the regression model and making predictions to evaluate accuracy. Linear regression is used, which relates price positively to rooms, address and area, but negatively to parking. The model has 70% accuracy in predicting relationships. Issues discussed include accuracy, transparency, fairness and privacy.