Housing-Price-Predictor

This project is a Linear Regression Model built to predict the sale price of residential homes using the Ames Housing Dataset. By leveraging a wide range of categorical and numerical features, the model uses Linear Regression to estimate housing prices with a focus on accuracy and interpretability

Model: Linear Regression

Dataset: Ames Housing Dataset

Preprocessing:

Dropped irrelevant features

Handled missing data with appropriate strategies (e.g., mean imputation, 'Not Available')

Applied normalization techniques such as z-score scaling and log transformation

One-hot encoded categorical variables

Feature Engineering:

Grouped and labeled categorical features

Scaled the target variable (sale price) for improved model performance

Evaluation:

Random Forest RMSE: 0.25795381307854687

Random Forest MAE: 0.18860455948688243

Random Forest R2 Score: 0.9434330669349795

Objective:

To build a clean and modular machine learning pipeline capable of accurately predicting home sale prices and generalizing well to unseen data.