QUESTION 1 The following table has 4 training instances and...

QUESTION 1 The following table has 4 training instances and 1 test instance. In naïve Bayes sentiment analysis (Positive or Negative), what is the prior for positive (i.e., $p(\text{positive})$)? 2 points Saved Reviews Sentiment Class Training Data Great movie. Positive I enjoyed it. Positive Not a big fan of the movie. Negative It is just average. Negative Test Data Enjoyed the movie. ? ? 1.00 ? 0.05 There is no need to calculate the probability for it. • 0.50 ? 0.00

Labs

Want to see this concept in action?

NEW

Explore this concept interactively to see how it behaves as you change inputs.

View Labs

Transcript

00:01 Hey there, welcome to numerade.

00:04 We are looking at a table here that summarizes the testing for a certain disease.

00:10 We're asked to find a probability that a subject does not have the disease, given that the test result is positive.

00:19 So again, what we can write down here is a base theorem equation, where we have the probability of does not have a disease.

00:30 So i'm going to denote this as no disease, given that the test result is positive.

00:47 So what this basically equals is the probability of getting a positive result, given no disease, multiplied by the probability of a positive result.

01:11 And then this is going to be divided by.

01:16 So what we get here is going to be divided by.

01:25 So this should be no disease.

01:46 Divided by the probability of getting positive results.

01:56 All right.

01:57 So what this basically will simplify into.

02:01 So this will equal the probability of a positive, well, sorry, a no disease.

02:22 And having a positive result.

02:27 So this thing is a bit too high.

02:36 And a positive result, this is then divided by the probability of a positive result.

03:06 All right.

03:08 So we can plug in our values here.

03:16 So in which we have for our first part, does not have the disease given the test result is positive...

Question

Please give Ace some feedback