Real-estate-Price-Prediction-
Project from course of Machine Learning
Real-estate accounts for more than 1.13 trillion of the USA economy and purchasing/selling a house is among the biggest commitments for most people. Accurate prediction of prices based on other sales can be a critical tool to make sure the buyer/seller is making an informed decision. While different realestate companies are working on custom algorithms to provide an estimate of the value of the property, they don’t provide insight into the method used and have varying degrees of error in different locations. The paper evaluates a combination of Stochastic Gradient Descent, Stochastic Dual Gradient Ascent, Gradient Tree Boosting and clustering using K-Means to predict real-estate prices. The input to the algorithm is data from Kaggle that has listing attributes and price for all the houses sold in King County from May 2014 to May 2015. The algorithm uses the attributes to train a model and predict the price.
ref. work - http://cs229.stanford.edu/proj2019aut/data/assignment_308832_raw/26646708.pdf