Geek Answers Handbook

Interactive decision tree classifier

Can anyone recommend the implementation of a classifier for the decision tree in Python or Java, which can be used gradually?

All implementations that I found require that you provide all the functions to the classifier immediately in order to get the classification. However, in my application, I have hundreds of functions, and some of the functions are functions that can take a long time to evaluate. Since not all branches of the tree can use all functions, it makes no sense to give the classifier all the functions at once. I would like the classifier to request functions, one at a time, in the order in which they are needed to minimize entropy and ensure the final classification.

+5
source share
5 answers

, , , .
, , " " , - , , ( , ). , . RF , , .

+2

( J48 Weka ) , , C4.5:

node , , , .

, , . (, ...) , " " node.

+2

? , , , , . . . , , . - .

, , , , , . , , . node, , .

, . , 3-5 , , . , .

, , , . , , .

0

, . , , - . , 20 .

"/", . , . , , , .

, , , :

Disease Name  Head Ache   Fever  Back Pain  Leg Pain  Blurry Vision  Hearing Loss
Common Cold   Yes         Yes    No         No        No             No
Migraine      Yes         No     No         No        Yes            No
Herpes        No          Yes    No         No        No             No

, , , .. , - . , .

  • , .
  • , .
  • , node, "/" ( Head Ache).
  • , .
  • , No.
  • 2

, , , , :

: blah blah blah.

1 , . , / . 1 / . , , , .

0

The Python decision tree package at https://pypi.python.org/pypi/DecisionTree has an interactive mode for using the decision tree. This is not incremental in the sense that you need. However, you can easily change the code in the function for an interactive operation so that you can see the results step by step.

0
source

More articles:


All Articles