House Price Prediction
Capstone Presentation
P R EPA R ED BY
VA I BH AV K U M A R
Introduction to problem statement
•A house value is simply more than location and square footage.
•Like the features that make up a person, an educated party would want to know
all aspects that give a house its value.
•For example, if we want to sell a house and we don’t know the price which we
may expect — it can’t be too low or too high.
•To find house price we usually try to find similar properties in our neighborhood
and based on gathered data will try to assess the house price.
•The problems faced during buying a house viz., buying a house is a stressful
thing, buyers are generally not aware of the factors that influence the house
and hence usually we take help of real estate agent which acts as a middleman
between buyer and seller and this resulted increased in cost of the house.
Understanding of dataset
This dataset has 21613 numbers of rows and 23 columns and the
details of attributes as given below:
Modelling approach used & why
•We have applied two algorithms for prediction of house price on the
given dataset viz., Linear Regression and Random Forest
•The reason for this is because linear regression helps in predicting the
future based on the past relationship of variables.
•In the given dataset we have only one predictor variable so therefore
linear regression has to be applied for model building.
•We have also used Random Forest classifier (which is in fact Random
Forest regression) it is possible to predict the house prices fairly good.
Modelling approach used and
why…contd
Modelling approach used and why…
contd
•We have used Gradient Boosting Regressor because it has fairly high R2 score as this algorithm
has advantage of increasing the algorithm's efficiency by minimizing over fitness.
•We have tuned our model and after tuning the R2 further improve as given in the below metrics
Plotting the tuned model
Model Interpretation
Model Interpretation… contd
In the below exhibit the red line shows the perfect predictions. If the prediction would equal the real value,
then all points would lie on the red line. Here you can see that there are some deviations and a few outliers, but
that is mainly the case for prices which are extremely high. There are some outliers in the low range and it
would be interesting to find out what is the cause of these outliers.
Business Insights and Recommendations
o Housing prices are an important reflection of the economy, and housing price
ranges are of great interest for both buyers and sellers.
o Therefore, according to our analysis, number of bedrooms, number of bathrooms,
square footage total floors have the greatest statistical significance in predicting a
house’s sale price.
o In terms of Business-to-Business segment, the prices of the house in the particular
project of the builder depends upon quality of construction, and furnished or not.
Because in metro and urban cities people would more likely to buy ready to move
house based on these parameters.
o In terms of Business to Consumer, the prices of the house would be driven by the
total area, covered area, number of floors, number of bedrooms, number of
bathrooms and availability of basement as per the needs of the prospective buyer.
o As housing market is not immune to wider political issues and economic forces
so therefore developers would try to build the houses more in the price band of
Rs.4Lacs – Rs.8Lacs would that would be a attractive segment as per the given
dataset.
o Similarly, eye-watering price for a three-bedroom property is more attractive
to the buyer instead of other segments. Therefore, developers would try to
accumulate 3BHK dwelling units more in comparison to others as per given
dataset.
o Further, as per the given dataset the average price of the 3BHK house that
prospective customer would like to buy is something around 3Lacs, but in case
of amenities like, in frontage of waterfront and sight this price can be stretched
up to ~3.5Lacs which a prospective consumer would be happily to buy.
o However, for first-time buyers are often looked as key to the buoyancy of the
housing market this may be a fascinating criterion whilst considering the price
of the house but for the second-time buyers who are looking for the
investment criteria this may not be an attractive criterion.
Thank You
