Rediscretize Continuous Nodes
Context
- Rediscretize Continuous Nodes is an option when setting up a Structural Coefficient Analysis.
- When Rediscretize Continuous Nodes is set, it performs an automatic rediscretization for each iteration, i.e., for each value of the Structural Coefficient, on the nodes that were originally discretized with an automatic discretization algorithm.
- Nodes that were discretized manually are not included in the rediscretization. Such nodes keep their original intervals throughout all iterations.
- Rediscretize Continuous Nodes uses the originally selected discretization algorithm and the originally selected number of intervals.
Tree-Based Discretization Algorithms
- Rediscretize Continuous Nodes is particularly useful in the context of Supervised Learning when used in conjunction with Tree-based discretization algorithms.
- Tree-based discretization algorithms utilize the Minimum Description Length Score, which depends directly on the Structural Coefficient.
- As a result, the discretization thresholds are optimized for each value of the Structural Coefficient.
- This approach can help you fine-tune a model in terms of trading off fit and complexity.
- Note that the Target Node in a network is never rediscretized automatically in this context.
Stochastic Discretization Algorithms
- If a Fixed Seed is set for the Random Number Generator under Main Menu > Window > Preferences > General > Random Number Generator, the following stochastic discretization algorithms would always produce identical discretizations each time they are run:
- Only if a Fixed Seed is not set, using the option Rediscretize Continuous Nodes will utilize the stochastic nature of the above discretization algorithms and produce stochastic variations in the discretization results.