2
$\begingroup$

I read two papers discussing how to compare between decision trees. One of them is RTED (that is specifically for trees and sub-case is decision trees) can be found here (RTED: A Robust Algorithm for the Tree Edit Distance)here:
and additional approach is COMBINING CLASSIFICATION TREES USING MLE by Shannon and Banks (Shannon, W. D., & Banks, D. (1999). Combining classification trees using MLE. Statistics in Medicine, 18(6), 727-740.). I would like to know if the MLE method is a sub case of RTED algorithm for comparison between decision trees? "Comparison between decision trees" I mean, given 2 decision trees we would like to compare the structure of them. It means by the location and the name of each node. The RTED transfers the trees into a Bracket Tree Format that contains the names of the tree nodes and compares between the 2 strings. The output is a number that is the Tree distance value (also know in literature Edit Distance). The question is if the algorithm described by Shannon and Banks (called MLE) is a specific case that is covered by RTED algorithm. If the final result of comparison between 2 decision trees will yield same result. If the answer is No, what are the cases in which the different algorithms will give different results?

$\endgroup$
3
  • $\begingroup$ Welcome to CS.SE! What do you mean by "comparison between decision trees" and by "sub case"? What are your thoughts, and what's your reasoning? I suggest editing the question. It'd also help if you could give a self-contained summary of those algorithms, if they're simple enough. Also, you can clean up the citations a bit: We expect references to fulfill the minimal scholarly requirements and be as robust over time as possible. Please take some time to improve your post in this regard. We have collected some advice here. Thank you! $\endgroup$
    – D.W.
    Commented Sep 13, 2016 at 21:02
  • $\begingroup$ Rather than leaving clarifications in the comments, please make sure to edit the question, and edit it into a form that reads well for someone who encounters the question for the first time. Right now the question is a bit of a wall of text. Would you like to edit the question to make this a bit more friendly to read? $\endgroup$
    – D.W.
    Commented Sep 13, 2016 at 21:32
  • $\begingroup$ I edited the question as you guided me with your clever questions. Can you help me please with further editing? $\endgroup$
    – Avi
    Commented Sep 13, 2016 at 21:35

1 Answer 1

1
$\begingroup$

The main difference between MLE and RTED is that RTED can compare any kind of decision tree even non binary such as CHAID. In addition, each of them has different approach for scanning the tree. MLE is from top down and find all the possible paths while RTED can go in breadth way.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.