Library
What is the difference between the Target Opimization and Target Dynamic Profile? They both appear aimed at optimizing some aspect of the target node and while the results obtained from each share a fair amount of similarity, yet they are not exactly the same. Under what circumstances should each be used?
Quote 0 0
Library
The Target Optimization is based on an exhaustive search algorithm returning the tuples that optimize the objective function specified by the user. Depending on the size of the network and the size of the tuples, the execution time can very quickly be prohibitive.targetOptimization.jpeg The Target Dynamic Profile is based on a heuristic search algorithm that looks for a set of “drivers” to optimize the same objective function. However, cannot be guaranteed that the returned solution is optimal.targetOptimization1.jpeg 
Quote 0 0
Pejokkes
What is the score and how to interpret it what can be seen in Target Optimization Tree results and in each row of its report?
Quote 0 0
Dan
The score in the Target Optimization Tree depends on the settings you have used. For example, if you are maximizing the Target Mean, the score of a leaf is the Target Mean associated with the leaf - the one associated with the root node (the marginal one). When you are maximizing the Target Mean by taking into account the joint probability, the score of a leaf is the Target Mean associated with the leaf - the one associated with the root node (the marginal one) x the probability of the joint observations that have generated the leaf.The Report computes, for each node-observation, the sum of the scores associated with all its leaves.
Quote 0 0