This project seeks to use data science techniques to discover whether building more homes has an impact on sales prices and the affordability of housing. The project gathers data from UK government sources, creating a dataset of Local Authority data for the years 2002 to 2021. A neural network is used to set benchmark predictive capability on the new data set. Alternative models that offer interpretability of the important features within the dataset are then employed. Aiming to discover if the measures that account for adding more housing contribute to the prediction, including what impact that has on prices. An alternative approach to measuring affordability is offered based on fuzzy logic. The results show that the available measure for adding new properties to the market has little impact on the prediction of prices to contribute to better affordability. Income, population and economic factors are more important.
This repository has been archived by the owner on Dec 9, 2022. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 0
MichelleBark/MSc-Project---House-Price-Prediction
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
Using data science techniques to discover whether building more homes has an impact on sales prices and the affordability of housing
Topics
machine-learning
data-mining
neural-network
tensorflow
scikit-learn
exploratory-data-analysis
keras
regression
pandas
data-visualization
seaborn
feature-selection
matplotlib
predictive-modeling
support-vector-machines
fuzzy-logic
model-evaluation
house-price-prediction
feature-importance
extra-tree-regressor
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published