5 Questions
What is the appropriate method to fill in the NaN’s present in a column with categorical variables?
Mode
What is the probability of failure of an air conditioning system due to a gas leak?
55/107
Given evidence of a gas leak in the failure of an air-conditioning system, what is the probability that electrical failure could have occurred?
55/72
Given evidence of electrical failure in an air-conditioning system, what is the probability that a gas leak could also have occurred?
55/107
Which metric is used for the continuous output problem?
R-square
Study Notes
Dealing with NaN's in Categorical Variables
- For categorical variables, it's not suitable to fill NaN's with mean or median values, as it may not make sense in the context of the data.
- Instead, possible approaches include:
- Removing rows with NaN values
- Replacing NaN values with a new category (e.g., "Unknown")
- Using a model that can handle missing values (e.g., decision trees)
Probabilities in Air Conditioning System Failure
- The probability of an air conditioning system failure due to a gas leak is unknown.
- Given evidence of a gas leak, the probability of electrical failure occurring is unknown.
- Given evidence of electrical failure, the probability of a gas leak also occurring is unknown.
Metrics for Continuous Output Problems
- For continuous output problems, a suitable metric is needed to evaluate the model's performance.
- Possible metrics include Mean Squared Error (MSE), Mean Absolute Error (MAE), and R-Squared.
Test your knowledge of data science and AI concepts with this Winter 2022 review quiz. From handling missing data in categorical variables to understanding the probability of failure in air conditioning systems, these review questions will help reinforce your understanding of key topics.
Make Your Own Quizzes and Flashcards
Convert your notes into interactive study material.
Get started for free
