Welcome! Log In Create A New Profile

Advanced

Decision trees

Posted by AliWasp 
Announcements Last Post
Announcement SoC Curricula 09/30/2017 01:08PM
Announcement Demarcation or scoping of examinations and assessment 02/13/2017 07:59AM
Announcement School of Computing Short Learning Programmes 11/24/2014 08:37AM
Announcement Unisa contact information 07/28/2011 01:28PM
Decision trees
November 03, 2010 09:37AM
Hey can some one help me out with working out the gain with regards to decision trees, please. I'm looking at the exam paper example for 2010, question 5.2, How do we get firstly the 0,1 for the first I and then how do we go from the fractions in brackets to the next line???
ie how does 5/14 I(3/5,2/5) end up being 5/14.0.98??
I think I'm being really doff sad smiley
avatar Re: Decision trees
November 03, 2010 09:55AM
The 0 and 1 in the first line is because there are 4 entries for Overcast, and all four of them have a YES Outlook. 0/4 = 0 & 4/4 = 1.

Then you have 5/4 * I(3/5, 2/5), you take the p value which is 3/5 = 0.6 go to the table in appendix A (at the end of the tut) find 0.6 and it gives you the value for I, which is 0.98.
Re: Decision trees
November 03, 2010 10:01AM
Ah ha! I see now!! Thank you so much I had figured out the p value part!! smiling smiley
Sorry, only registered users may post in this forum.

Click here to login