Class Prediction - Data analytics

27 views Asked by At

I have 2 sets of data, training and testing dataset. I need to predict a class(ExpiredHospital) for the testing data. I need to train the model so that it would give me a good predicted value for ExpiredHospital.

I tried using the K-Nearest Neighbor method with knime and it give me a result of 0.74186 in kaggle which contains the actual value of ExpiredHospital. I aim to get a higher score in kaggle. What configuration should I use and what data preprocessing can be used for this. Correlation between attributes. Here is the configuration for KNN node that I am using

KNN node configuration

And here is my workflow:

KNIME workflow screenshot

What other method should I use and the configuration. The testing model has 11796 rows

0

There are 0 answers