/House-Price-Prediction

House Price Prediction Assignment

Primary LanguageJupyter Notebook

House Prediction With Ridge and Lasso Regression

Table of Contents

General Information

  • This task will predict using two models Ridge and Lasso Regression
  • This will help business to determine which house to buy
  • A US-based housing company named Surprise Housing has decided to enter the Australian market. The company uses data analytics to purchase houses at a price below their actual values and flip them on at a higher price. For the same purpose, the company has collected a data set from the sale of houses in Australia.
  • We have a training dataset added with the code which have all features for the House and based on these features are needed to make selection for which property to buy

Conclusions

  • After running the models we came into conclusion that we can use Ridge over Lasso because it has better r^2
  • These are the important predictive variables
  • LotArea------------- Lot size in square feet
  • OverallQual--------Rates the overall material and finish of the house
  • OverallCond-------Rates the overall condition of the house
  • YearBuilt-------- ---Original construction date
  • BsmtFinSF1-------Type 1 finished square feet
  • TotalBsmtSF------Total square feet of basement area
  • GrLivArea----------Above grade (ground) living area square feet
  • TotRmsAbvGrd---Total rooms above grade (does not include bathrooms)
  • Street_Pave-------Pave road access to property
  • RoofMatl_Metal--Roof material_Metal

R^2 of Ridge and Lasso for Train and Test dataset

  • Ridge
  • R2 score(Train)--------- 0.88 ----------------------------0.88
  • Lasso
  • R2 score(Test)-----------0.87-----------------------------0.86

Technologies Used

  • Pandas - version 1.4.3
  • Numpy - version 1.23.2
  • Sklearn - version 1.1.2
  • Seaborn - version 0.11.2
  • Matplotlib - version 3.5.3

Acknowledgements

  • This project was inspired by Upgrad Tutorials
  • Refrerences - Kaggle, Sklearn Documentation

Contact

Created by [@sudeepignition] - feel free to contact me!