System and method for managing routing of customer calls to agents
摘要
說明書
In an embodiment, the tree-based model 418 is a random forests model. Random forests is a class of ensemble methods used for classification problems. Random forests models work by fitting an ensemble of decision tree classifiers on sub samples of the data. Each tree only sees a portion of the data, drawing samples of equal size with replacement. Each tree can use only a limited number of features. By averaging the output of classification across the ensemble, the random forests model can limit over-fitting that might otherwise occur in a decision tree model.
In an embodiment, the tree-based model 418 uses the random forests model in Python's scikit-learn. In an illustrative embodiment, the tree-based model 418 uses the following parameters in the scikit-learn random forests model:
-
- Maximum tree depth: 3 or ∞, set with max_depth.
- Maximum number of features considered when looking for the best split: 3→6, set with max_features.
- Minimum number of samples required to split a node of the tree: 2→11, set with min_samples_split.
- Minimum number of samples to be a leaf node: 1→11, set with min_samples_leaf.
- Number of trees in the forest: 100 or 200, set by n_estimators.
- Whether to sample with replacement for the data seen by each tree: true or false, set by bootstrap.
- Function to measure quality of a split: Gini or Entropy (information gain), set as criterion.
