Predict House Price in King County with Azure Machine Learning and Power BI (Part 3)
So in the part 1 & 2, I have explained about Regression Analysis and clustering. In Part 3, I will explain about visualization with Power BI.
You can read Part 1 & 2 through this link:
Predict House Price in King County with Azure Machine Learning and Power BI (Part 1)
Predict House Price in King County with Azure Machine Learning and Power BI (Part 2)
King County, Washington, United States is the most populous county in Washington. A large number of residents attracts property sellers in this area. So in this Microsoft X Studi Independen Kampus Merdeka Capstone project, I take the case of MariBisnis, a company that wants to know the house price prediction in King County.
To solve this problem, Microsoft provides resources with less/no code to make predictions and applications to visualize.
Purpose
- Train machines with existing data in order to make house price predictions.
- With the existing data, it can be seen the trend of home sales from time to time as information on home sales business strategies
- The existing data will be visualized so that the data can be understood by the managerial level.
Benefits
- Using visually displayed analysis can summarize a picture of the current state of the property
- The analyzed data will then be used for future decision-making.
Visualization Using Power BI
You can check my Power BI Report through this link
House Category Visualization
- The house that has the most sales and has the highest sales is the Middle Price House type.
- Sales from month to month were quite stable and house additions were quite stable during this period.
Price Based On Location
Areas with postal code 98115 have high total sales and all types of homes sold are Low Middle Price.
Price Based On Quality
- If a house has a waterfront, it tends to be more expensive
- The higher the grade of a house, the more expensive it is
- The better the view of a house, the more expensive it is
Price Based On Core Facilities
The house with the highest average price has 8 rooms, 7.75 bathrooms, and 2.5 floors.
Price Based On Area
The larger the land area, especially the basement, plots, and houses, the more expensive the house will be.
Trend Price Prediction
Housing prices are predicted to decline further in the future.
Conclusion from All Part
• The best model for performing regression analysis is boosted decision tree regression using hyperparameter tuning
• MAE Using the boosted decision tree regression model Using hyperparameter tuning is 6.3950e+4
• The most sold house category is Middle Price House
• The housing category with the highest sales value is Middle Price House with a total of 4 billion
• The development of house prices in each category tends to be stable.
• Although the areas with the postal code 98115 are all Low Middle Price houses, home sales in that area are the highest.
• The more rooms, the waterfront, and the higher grade, the higher the selling price of the house
• Houses with high prices on average have facilities with 8 rooms, 7.75 bathrooms, and 2.5 floors.
• The wider the land/basement/house, the more expensive the house price.
• Based on the predicted price, in the future, sales will decline.
Recommendations
• Because the model and time are very limited in conducting regression analysis, the score obtained cannot be said to be good.
• Perform additional information such as a description of each attribute.
• Perform additional data to improve model performance.
• Improvise by implementing other forms of models because the models available in Azure Machine Learning are very limited.
• Increase sales in the Low — Middle House Price category.
- Exploring other factors causing price declines.
Thank you for reading!
