Iam so confused using weka to my thesis. I want ro running data with weka but
still confused what can i use with my data. I want to know determinant
rating of bond. my dependent variable is rating AAA,A+,A,A-,BB+, BB, D. my
independent variable are leverage (numeric), liquidity(numeric) asset
(numeric), guarante (nominal), structure (nominal). What should i do for my
data ? Using multilayer perception or j48 but if i using them. I dont know
how to read they result. Please help me 😢 i want to graduate
> Thanks sir. Anyway can i run my data with neural network or j48 ? Which one
> do you choose ?
You can run any algorithm that handles classification on your data.
However, you will have to choose the best one through running
experiments. There is no easy answer to that, which is the reason for
Auto-ML (exploring algorithms and their parameter spaces).
J48 and RandomForest are always sensible choices to start with and use
ZeroR (which only predicts the majority class) to determine the
baseline that other algorithms need to beat.