Happy Wednesday, folks! We just posted a fresh Just KNIME It! challenge.
This week you’ll wear the hat of a real estate agent who just moved to a new city. Before directly working with potential buyers, she decided to first dive deep into the region’s data to better understand the market. What are her main insights?
Here is the challenge. Let’s use this thread to post our solutions to it, which should be uploaded to your public KNIME Hub spaces with tag JKISeason3-3.
Need help with tags? To add tag JKISeason3-3 to your workflow, go to the description panel in KNIME Analytics Platform, click the pencil to edit it, and you will see the option for adding tags right there. Let us know if you have any problems!
This afternoon Jakarta time, I have completed the solution to the given challenge. In tackling this challenge, I took on the role of a data analyst working at a real estate company. My solution can be accessed through this link, JKISeason3-3 arief_rama – KNIME Community Hub
I have created an interactive dashboard to check the correlation between price and housing features. The scatter plot shown below was created using Plotly and scikit-learn.
Hi all,
Here’s my solution, similar approach to @sryu. Random forest regression was used simply with default settings. The global feature importance shows that zipcode and sqft_living contribute to housing prices significantly.
House Prices JKI3-3 Leo_Wynter
Interesting challenge.
Got to try out some EDA followed by a bit of some machine learning (barebones, not much fixing)
Used the Statistics View for summary statistics, a histogram plot and Plotly’s violin plot node for checking price distribution as well as a heatmap for correlation.
Also played around with Palladian Map viewer (these homes were in Seattle?)
Had fun!!
I built a data app that, when first opened shows some filters in the form of range sliders, a button to visualize the houses you select and a data table of the different real estate listings:
When you use the ranger sliders to narrow the listings, the table below is automatically updated. You can then sort using the table and finally, using the check boxes, select a few that you are interested in learning more about:
Now you hit the “Visualize Selected Houses” button and presto! A map is displayed showing you the location of each house you selected from the table (blue arrows)!
If you zoom in on one of the listings, you see that there is more information available. For each house selected, the Haversine distance to all other listings is calculated and anything within a certain threshold is also shown on the map. Additionally, the OSM point of interest nodes were used to add schools and fuel stations to the map, since these are possible amenities one would consider in buying a house. Others could be added!!
Happy Wednesday, folks! We just posted a fresh Just KNIME It! challenge. 
This week you’ll wear the hat of a real estate agent who just moved to a new city. Before directly working with potential buyers, she decided to first dive deep into the region’s data to better understand the market. 
What are her main insights?
Need help with tags? To add tag JKISeason3-3 to your workflow, go to the description panel in KNIME Analytics Platform, click the pencil to edit it, and you will see the option for adding tags right there. 
Let us know if you have any problems!
Excellent, in my opinion, this is exactly the approach I was thinking of to tackle this challenge: Predict House Price ==> Features Importance.
