Tutor Hunt Questions
“can someone help me with this question In your data base you have N examples from the first class (“good”) and M examples
from the second class (“bad”). In real life we know that the frequency of good cases is q and the
frequency of bad cases is 1-q (the proportion of good and bad cases in your data base may be different
from the real life). Let the kNN classifier give you k nearest neighbours from your data base for a
given case with unknown class. From these k nearest neighbours v are of the good class and w=k-v of
the bad class. How will you evaluate the probability that your example is “good”?
from the second class (“bad”). In real life we know that the frequency of good cases is q and the
frequency of bad cases is 1-q (the proportion of good and bad cases in your data base may be different
from the real life). Let the kNN classifier give you k nearest neighbours from your data base for a
given case with unknown class. From these k nearest neighbours v are of the good class and w=k-v of
the bad class. How will you evaluate the probability that your example is “good”?
7 years ago
Maths Question asked by Khaznah
Add a CommentKnow the Answer?
Please enter your response to the question below. The student will get a notification as soon your response has been approved by our moderation team.
1 Answer
The probality that it is good depends on the initial listed values
Answered by [Deleted Member]
- 6 
