Description
Overview
Creating a predictive model is a very common activity in artificial intelligence. In this lab, you apply Naïve Bayes
inference to the Kaggle adult Data Set, which asks you to predict whether income exceeds $50,000 per year
based on census data.
Step 1 – Download Files
Download the lab9.py file and the adults.csv file from D2L and place them in the same folder.
The lab9.py file uses the following libraries:
• numpy
• pandas
• matplotlib.pyplot as plt # for data visualizations
• sklearn.naive_bayes.GaussianNB
• seaborn
• warnings
Run the lab9.py program to ensure all of the libraries work, and the program loads the CSV file correctly. Others
may be added along the way, but this will get you started.
Step 2 – Tutorial
Complete the remainder of the tutorial on the following site, beginning with Step 7 and STOP at the end of Step
15:
• https://www.kaggle.com/prashant111/naive-bayes-classifier-in-python
Step 3 – Questions
Answer the following questions in Lab9.docx.
1. Explain why feature engineering was necessary in this problem.
2. Explain why feature scaling was necessary in this problem.
3. Does this model have a good accuracy? Why or why not?
4. Speculate about ways to improve the accuracy of the prediction.
5. Given these results, and referring to this problem specifically, explain why some use the term “idiot
Bayes.”
CSCI 5260 – Artificial Intelligence Page 2
Submission
Create a Word Document named Lab9.docx with your responses to the questions. Submit your complete
lab9.py file.
Submit to the Lab 9 dropbox at or before Monday, April 5, 2021 by 11:59 PM.
Grading
A letter grade will be assigned for each response. The letter grades are based on both correctness and the
adequacy of answers. Points are assigned as follows:
A B C D F Zero
Excellent Above
Average Average Below
Average Poor No Attempt
10 8 6 4 2 0
Naïve Bayes
Classification
Step 2
Step 3-
1
Step 3-
2
Step 3-
3
Step 3-
4
Step 3-
5
