Data Mining with Weka (4.3: Classification by regression)

[@JohnLee-dp8ey]

4:30 In the current stable version of WEKA (Version 3.8.6), you have to choose PlainText for Output Predictions to output the prediction table

[@CoolDude911]

You train the regression model on the entire dataset and save that as an attribute. So when you do cross validation for the One R classifier then you are using an attribute for a model that was trained on the entire dataset (including the test set for that fold). Isn't that a problem?

[@terrencetran0]

Hi thank you for this amazing video, I have a question. How can we save the model to use it on new data ? For example, I have a new patient and all the attributes, how can I use the model to tell if he is Positive or Negative ?

[@user-sk3dr8nl9u]

thank you

[@xcriss2898]

Excellent trick using the linear regression here. Thx!

[@FarizDarari]

The lecture did give a suitable abstraction for the students to learn Weka :)

[@mpdota241]

thank you sir for the lesson,, relly helped me