House Price Prediction

This repository contains the code for a house price prediction project. The goal is to develop a regression model to predict house prices based on various features.

Overview

• HousePricePrediction.ipynb: Jupyter Notebook containing the main code for data analysis, preprocessing, model training, and evaluation.
• HousePricePrediction.xlsx: Dataset used for training and testing the models.

Project Structure

• Data Exploration and Preprocessing: The notebook begins with an exploration of the dataset, identifying and handling categorical variables, and cleaning missing values.

• Exploratory Data Analysis (EDA): Various aspects of the dataset are analyzed, including correlations between numeric features and the distribution of categorical features.

• Data Cleaning: The 'Id' column is dropped, and missing values in the 'SalePrice' column are filled with the mean. Rows with missing values are removed to create a clean dataset.

• One-Hot Encoding: Categorical variables are one-hot encoded using the OneHotEncoder from scikit-learn.

• Splitting Dataset: The dataset is split into training and validation sets for model training and evaluation.

• Modeling and Evaluation:

  • Support Vector Machine (SVM)
  • Random Forest Regressor
  • Linear Regression
  • CatBoost Regressor

Dependencies

• Python 3
• Jupyter Notebooks
• Required Python packages: pandas, matplotlib, seaborn, scikit-learn, catboost

Usage

1. Clone the repository:

git clone https://github.com/frtpynrc/House_Price_Prediction.git