Explore BrainMass
Share

Explore BrainMass

    Tree Pruning to avoid data overfitting

    This content was COPIED from BrainMass.com - View the original, and get the already-completed solution here!

    "Tree Pruning." Our textbook says that the tree pruning addresses the problem of "overfitting the data." Discuss how overfitting affects the performance of a decision tree.

    Some overview:
    http://chem-eng.utoronto.ca/~datamining/dmc/decision_tree_overfitting.htm

    © BrainMass Inc. brainmass.com October 10, 2019, 4:35 am ad1c9bdddf
    https://brainmass.com/engineering/mining-engineering/tree-pruning-avoid-data-overfitting-468291

    Solution Preview

    Hello,

    A problem that arises with all methods of generating classification rules is that of overfitting to the training data. In some cases this can result in excessively large rule sets and/or rules with very low predictive power for the unseen data. A smaller number of more general rules may have greater predictive accuracy ...

    Solution Summary

    This solution provides concept on how to avoid over fitting in data mining by utilizing tree pruning. Pre pruning and post pruning are used for this purpose.

    $2.19