Help with multi-label classification

[rogerius1st]

Hello, dear Knimers,
I have another issue: I made a survey with several professionals, asking questions about their business situation over a series of months. I investigated three different variables (COVID-related rates) as potential determinants of bad performances of their businesses. These questions had six different alternatives as possible answers. These options are categorical ordinal variables, with string values, representing percentage intervals (e.g., <5%; 5-10%;… ; >75%). I have already converted them to other formats for labeling each value, both as string and integer values.
Now, I wish to know if there is (and which are) the algorithms available in Knime for the desired multi-class classification.
Thanks for any help.
B.R.,
Rogério.

[mlauber71]

@rogerius1st you could check the links about multiclass in my machine learning collection

Machine Learning Meta Collection (with KNIME) – mlauber71

And then these sample workflows:

https://hub.knime.com/search?type=Workflow&tag=multiclass,h2o.ai&sort=maxKudos

A good statistic to go for might be LogLoss:

Any advice to improve the performance of a classification model ?

We had a similar discussion about metrics in multi class predictions here: Besides accuracy you could use a LogLoss metric. That means that not only the outcome of prediction will be taken into account but also the ‘level of certainty’. If you have two predictions from two models A) 1=51%, 2=39%, 3=10% TRUTH=>1 B) 1=90%, 2=5%, 3=4% TRUTH=>1 both come up with the correct result (1). But prediction B) would be better as it predicts the correct answer with higher confidence and might be mor…