Finding an Optimal Classification based on Cost and Profit
This workflow trains a classification model using the Linear Regression algorithm. The class prediction is evaluated by overall accuracy and profit statistics. To calculate the profit statistics, classification results are weighted by the values in a predefined profit matrix. The values in the profit matrix depend on the final goal of the classification model.
URL: How to Optimize the Classification Threshold https://www.knime.com/blog/from-modeling-to-scoring
URL: Scoring Metrics eBook https://www.knime.com/knimepress/scoring-metrics-evaluating-machine-learning-models
Nodes
- Line Plot4 ×
- Column Filter3 ×
- Constant Value Column2 ×
- Rule Engine2 ×
- Column Expressions1 ×
- Show all 23 nodes
Extensions
- FeatureKNIME Base nodes
- FeatureKNIME Column Expressions (legacy)
- FeatureKNIME JavaScript Views (Labs)
- FeatureKNIME Javasnippet
- FeatureKNIME Views
Links
- No links available