Icon

20230622 Pikairos JustKNIMEIt Season 2 Challenge 13 Understanding Crime and Real Estate Connections

20230622 Pikairos JustKNIMEIt Season 2 Challenge 13

You ara a data scientist working for a real estate company, and heard a rumour that the "average number of rooms per dwelling" (RM) may be connected to the "per capita crime rate" (CRIM) depending on the city/town. You then decide to investigate if this is the case for Boston, the city where you live and work from. To this end, you decide to experiment with a machine learning regression model and with a topic that you have recently been studying: XAI. How are RM and CRIM connected in Boston? Hint: Consider calculating the SHAP values of each independent feature using a SHAP loop. Hint 2: Consider using a dependence plot to verify how RM and CRIM are connected visually.

Challenge 13You ara a data scientist working for a real estate company, and heard a rumour that the "average number of rooms perdwelling" (RM) may be connected to the "per capita crime rate" (CRIM) depending on the city/town. You then decide toinvestigate if this is the case for Boston, the city where you live and work from. To this end, you decide to experiment with amachine learning regression model and with a topic that you have recently been studying: XAI. How are RM and CRIMconnected in Boston? Hint: Consider calculating the SHAP values of each independent feature using a SHAP loop. Hint 2:Consider using a dependence plot to verify how RM and CRIM are connected visually. HousingDataPort 0 : Train SetPort 1: Test Set SHAP : Look at the viewof this componentLoop EndLoop EndLoop StartLoop Start UsesSHAP Summarizer Sampling weightSampled 100 rowsTrain ModelApply Modelto Predict TargetColumnUse k-means to Summarize the Data to n Prototype RowsApply Modelto Predict TargetColumnShapley Values : Look at the viewof this componentHeatmapBased on OriginalDataCSV Reader Partitioning Dependence Plot Shapley ValuesLoop End SHAP Loop End Shapley ValuesLoop Start SHAP Loop Start Row Sampling Post-Processing Gradient Boosted TreesLearner (Regression) Gradient Boosted TreesPredictor (Regression) SHAP Summarizer Gradient Boosted TreesPredictor (Regression) Dependence Plot Linear CorrelationHeatmap Challenge 13You ara a data scientist working for a real estate company, and heard a rumour that the "average number of rooms perdwelling" (RM) may be connected to the "per capita crime rate" (CRIM) depending on the city/town. You then decide toinvestigate if this is the case for Boston, the city where you live and work from. To this end, you decide to experiment with amachine learning regression model and with a topic that you have recently been studying: XAI. How are RM and CRIMconnected in Boston? Hint: Consider calculating the SHAP values of each independent feature using a SHAP loop. Hint 2:Consider using a dependence plot to verify how RM and CRIM are connected visually. HousingDataPort 0 : Train SetPort 1: Test SetSHAP : Look at the viewof this componentLoop EndLoop EndLoop StartLoop Start UsesSHAP Summarizer Sampling weightSampled 100 rowsTrain ModelApply Modelto Predict TargetColumnUse k-means to Summarize the Data to n Prototype RowsApply Modelto Predict TargetColumnShapley Values : Look at the viewof this componentHeatmapBased on OriginalDataCSV Reader Partitioning Dependence Plot Shapley ValuesLoop End SHAP Loop End Shapley ValuesLoop Start SHAP Loop Start Row Sampling Post-Processing Gradient Boosted TreesLearner (Regression) Gradient Boosted TreesPredictor (Regression) SHAP Summarizer Gradient Boosted TreesPredictor (Regression) Dependence Plot Linear CorrelationHeatmap

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: LAMBERT
Created at: 2023-06-22
On NodePit since: 2023-06-28
Last update: 2024-11-21
Created with KNIME version: v4.7.4
Tags: JKISeason2-13

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!