Due to recent updates, all users are required to create an Altair One account to login to the RapidMiner community. Click the Register button to create your account using the same email that you have previously used to login to the RapidMiner community. This will ensure that any previously created content will be synced to your Altair One account. Once you login, you will be asked to provide a username that identifies you to other Community users. Email us at Community with questions.

The importance of factors in predictive models

misdrymisdry Member Posts: 1 Learner I
in Help

Hello, teachers! I'm Misdry, beginner to RM. I have a question about how to explain the importance of the factors in the regression prediction model. For example, I have a 60W row and 18 columns of data, in which 17 columns are impact factors, which are all discrete 1-5 data, and the last column is the prediction label, which ranges from 0 to 1.

1. I established the neural network model and obtained the results through the simulation operator.I tried to adjust the value of each factor to be consistent, and observed the change of the predicted result by changing the value of one factor, for example, reducing the fourth factor from 5 to 4 to determine the change of the result. Is it reasonable to explain the importance of the factor in this way? Also, how are the simulated global weights calculated? Is there a better way?

2. Can the operator feature weights be used to evaluate the importance or influence of factors on prediction labels? Or can only be used as a step of factor selection before model training.

Thank you very much for reading my question patiently. I added XML at the end. I am looking forward to your reply! :)

Regards,


Sign In or Register to comment.