joyumoh/Nigeria-Real-Estate-Prediction

End-to-end real estate price prediction project using Nigerian housing market data. Open in Colab # Nigeria Real Estate Price Prediction ## Overview This project develops machine learning models to predict residential property prices in Nigeria using real estate data. The workflow covers data cleaning, exploratory data analysis (EDA), feature engineering, model training, evaluation, and feature importance analysis. The goal is to help investors, real estate professionals, and potential buyers estimate property values based on key property characteristics. --- ## Project Objectives * Analyze the Nigerian real estate market dataset. * Clean and preprocess property data. * Perform exploratory data analysis to uncover market trends. * Engineer features suitable for machine learning. * Train and compare multiple regression models. * Identify the most influential factors affecting property prices. * Build a foundation for automated real estate valuation systems. --- ## Dataset Features The dataset contains property-related attributes such as: * Property Size (Square Feet) * State * Location Type * Property Type * Property Condition * Annual Rent * Property Price (Target Variable) --- ## Project Workflow ### 1. Data Loading The dataset is imported and inspected using Pandas. ### 2. Data Cleaning * Checked for duplicate records. * Checked for missing values. * Formatted price and rent columns. * Detected and visualized outliers. ### 3. Outlier Treatment * Applied capping using the 99th percentile to reduce the impact of extreme property prices. ### 4. Exploratory Data Analysis (EDA) Performed visualizations to understand: * Price distribution * Annual rent distribution * Property size distribution * Average property prices by state * Relationships between property features and prices ### 5. Feature Engineering Categorical variables were transformed using One-Hot Encoding: * State * Location Type * Property Type * Property Condition ### 6. Feature Selection Prepared predictor variables (X) and target variable (y). ### 7. Train-Tes …