Question: Question 1. Decision Theory [7 Points] Trained classifiers may be used to determine the polarity of sentiments expressed in a tweet as Positive or Negative.
![Question 1. Decision Theory [7 Points] Trained classifiers may be used](https://dsd5zvtm8ll6.cloudfront.net/si.experts.images/questions/2024/10/6706912a596ef_6266706912a35c28.jpg)
Question 1. Decision Theory [7 Points] Trained classifiers may be used to determine the polarity of sentiments expressed in a tweet as Positive or Negative. A tweet may be inherently Complimentary or Derogatory. There are two available classifiers for this task: A and B. Sixty percent of the tweets are Complimentary. The performance of the classifiers are shown in the tables below: Classifier A Positive Negative Complimentary 0.50 0.10 0.60 Derogatory 0.04 0.36 0.40 Classifier B Positive Negative Complimentary 0.54 0.06 0.60 Derogatory 0.10 0.30 0.40 Assume that there are no costs when a Complimentary tweet is classified as Positive or a Derogatory tweet is classified as Negative. a. The cost when a Complimentary tweet is classified as Negative is $1, and the cost when a Derogatory tweet is classified as Positive is $2. Which classifier should a risk- neutral rational decision maker choose? Why? [3 Points]
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
