Scribd
Upload
550 views9 pages

Bangalore House Price Prediction Using The Best Machine Learning Model Submitted by Rukzana Vadakkekudy Rassak P2682221

Uploaded by

Adarsh Pj
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
550 views9 pages

Bangalore House Price Prediction Using The Best Machine Learning Model Submitted by Rukzana Vadakkekudy Rassak P2682221

Uploaded by

Adarsh Pj
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 9

BANGALORE HOUSE PRICE PREDICTION USING THE BEST MACHINE LEARNING

MODEL

SUBMITTED BY

RUKZANA VADAKKEKUDY RASSAK

P2682221
Contents
1.OBJECTIVE
2.LITERATURE REVIEW
3.METHODOLOGY
4.DATA VISUALISATION
5.MODEL SELECTION
6.CONCLUSION
7.HIGHLIGHTING THE CHANGES MADE
Objective
The main objective of the project :
To find the best machine model for predicting the house price for Bangalore city

 To find the best feature in predicting the house price.

Examine the performance of the model using the metrics r2 score ,mean square error
and mean absolute error
Literature Review

As a part of Literature review more than 17 articles of already implemented related works
were reviewed .Every work is different in conclusion.
The literature review helps to sort out the best models for house cost prediction.
From the review the models like SVM-Support Vector Machine Algorithm,Cat-Boost
Algorithm,Random Forest,Lasso Regression,Ridge Regression,Linear
Regression,Random forest,Decision tree and elastic net were selected for model
exploration.
Methodology
Python was used to construct the project, and Google Co-Lab provided the interface.
Steps involved in the Price Prediction using the data set.
1.Data-Preprocessing-includes importing libraries, data structure, data cleaning and feature
engineering.
2.Exploratory Data Analysis is done
3.Model Selection by checking its performance using the score and other three metrices
including,r2_score,mean absolute error and mean squared error
4.Feature importance is performed for models.
5.Cross validation is done to check the accuracy of best model.
Model selection
The Linear Regressions founds to be better in accuracy when compared to other top
models in the history of House price predictions.
The score level for Linear Regression is 0.86 which is better.
To measure the accuracy of Linear Regression we used k-fold cross validation. In the five
iterations, there was a score above 80%.
Performance table of Linear Regression is shown below

 Predictions on the different locations with bath, square feet, and bedroom have been done.
Conclusions
• The Linear Regression was found to be better in accuracy when compared to other top
models in the history of House price predictions.
• To measure the accuracy of Linear Regression we used k-fold cross validation. In the
five iterations, there was a score above 80%.
• From the heat map, the most relevant feature in the prediction of house price is the
Feature “total_sqft”.
• Checked the performance through r2_Score,Mean Squared Error and Mean Absolute
Error.
• Feature importance property of the model also performed.
• In the future, if more attributes need to be added to the dataset. The current information
from the data set is limited even though it is recently updated. People always consider
more features when purchasing a house like the direction which it faces, local shops, and
other facilities like educational institutions, banks, transport stations, etc which are not
included in the dataset.
HIGHLIGHTING THE CHANGES MADE

1.Added two more models.


2.Done the cross validation of best two models.
3.Added one more metrics to the existing one.
4.Implemented “feature importance”.
5.Created a table containing the performance of models using the metrices.
6.Updated the result and conclusions with the table of model performance and
comparison of feature importance of best two models.
7.Have gone through the feedback of both supervisors and done accordingly

You might also like