This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
|
||||||||
|
Paper Details
Paper Title
Improving Accuracy Of Real Estate Valuation Using Stacked Regression
Authors
  Dhvani Kansara,  Rashika Singh,  Deep Sanghavi,  Pratik Kanani
Abstract
Real Estate business is flourishing with each passing day making it imperative for an effective house prediction model. This in turn will be beneficial to both the investors as well as the estate owners who can get effective prices without depending on external third party agents or mere capitalization rates. Machine Learning can be leveraged for this purpose. This paper evaluates the algorithms which can be used to predict the house prices on the Boston dataset [1] taken from Kaggle with 79 attributes like Living area, Condition at time of sale, Proximity to roads and rails, year built, etc. taking into consideration all aspects from homes in Ames, Iowa. In order to predict House Price, which is the dependent variable, regression algorithms under supervised learning are used. This paper provides the most optimal solution with the Stacked Regressor, an ensemble model which in this case averages Multiple Linear regression, Random Forest Regression and XGBoost regression algorithms ultimately giving a Root Mean Square value of 0.1047 and a high accuracy of 93.52.
Keywords- Real Estate Price Prediction, Regression algorithms, Model Stacking
Publication Details
Unique Identification Number - IJEDR1803097Page Number(s) - 571-577Pubished in - Volume 6 | Issue 3 | September 2018DOI (Digital Object Identifier) -    Publisher - IJEDR (ISSN - 2321-9939)
Cite this Article
  Dhvani Kansara,  Rashika Singh,  Deep Sanghavi,  Pratik Kanani,   "Improving Accuracy Of Real Estate Valuation Using Stacked Regression", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.6, Issue 3, pp.571-577, September 2018, Available at :http://www.ijedr.org/papers/IJEDR1803097.pdf
Article Preview
|
|
||||||
|