Watch this video to see how to use logistic regression classifiers with publicly available data about metabolic diseases to diagnose chronic kidney disease.

For this video the Watson Machine Learning Apache Spark and object storage instances are already provisioned in IBM Bluemix, and there’s a project in Data Science Experience with all of the associated services. If you haven’t done so yet, watch the two “Getting Started” videos in the IBM Watson Machine Learning Learning Center:

Download and unzip the data file from the IBM data and analytics portal at
https://github.ibm.com/dap/dap-planning/files/13589/chronic_kidney_disease_file.csv.zip.

On the overview tab, in the File slide-out panel, browse for the CSV file and open it. Once it’s finished loading, you’ll see it in the list of data assets. Now you’re ready to create a new model. Name this model “Predictive Analytics” and provide a description.


The Machine Learning and Spark services are already associated. Select Manual for the training method and create the model.

After you load the data, you must train the data. This consists of choosing an appropriate technique and estimator to apply to the raw data. For this dataset, you’ll use the Class Label column and the Logistic Regression estimator type.

After you’re trained and saved the model, you’re ready to deploy it. Click Deploy.

When the model deployment is complete, view the deployment. Take note of the scoring End Point for future reference and remember that you can have only one deployment per user. Now test the model prediction using the sample input data and click Predict. This example shows that there is a one hundred percent chance that a 48-year-old with diabetes and a serum creatinine of 1.2 will have chronic kidney disease. Change the input data to test different predictions.

Resources for you

Cognitive Computing
Create apps that accelerate, enhance, and scale the human experience.

Join The Discussion

Your email address will not be published. Required fields are marked *