🦉 🎤   RapidMiner Wisdom 2020 - CALL FOR SPEAKERS   🦉 🎤
We are inviting all community members to submit proposals to speak at Wisdom 2020 in Boston.
Whether it's a cool RapidMiner trick or a use case implementation, we want to see what you have.
Form link is below and deadline for submissions is November 15. See you in Boston!
please help understanding confidence & prediction scores for vector and regular linear regression.
The goal is to use regression for binominal 0/1 classifiction task and generate a model for deploy. Here are the points:
??? 1. Vector LR output is labeled data with preditcion score, and model formula. However, applying this formula to the data gives us different prediction scores than labeled output. Why this happen, how can I make labeled output prediction scores be the same as model formula?
??? 2. Regular LR gives us confidences as output. How this confidences are calculated? We can convert it to prediction score using Rescale Confidences operator, but resuling predictions will be strictly in range 0..1 and, therefore, different from prediction generated applying linear formula to our data. What Rescale Confidences operator does, how to understand its 'prediction' score?
??? 3. Finally, how to apply vector and regular LR models, if I'm getting different prediction scores in RapidMiner and after model formular apply?
Sorry for this very basic questions, however I couldn't find any related information of forums or RM manuals/books.