Skip to content

ML project to predict housing prices with linear regression and random forest

Notifications You must be signed in to change notification settings

neslihancesurr/housing-models

Repository files navigation

Housing price models

Data

  • novemberdata.json: the raw data scraped from HepsiEmlak. includes around 25.000 apartment listings in İstanbul.
  • cleaned_data_model: the data after the application of data_cleaning file.

Cleaning and Correlations

  • emlak-data-cleaning: switching to numerical values, getting rid of outliers
  • emlak-correlations: visualizations of relationships between features

Models

  • emlak-linear: a linear regression model built with scikit-learn (K-fold cross-validation + MAE, NAE + price prediction example)
  • emlak-polynomial: a polynomial regression model built with scikit-learn (K-fold cross-validation + MAE, NAE + price prediction example)
  • emlak-randomforest: a random forest model built with scikit-learn (K-fold cross-validation + MAE, NAE + price prediction example)

About

ML project to predict housing prices with linear regression and random forest

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published