MLS-C01 · Question #97
A Data Scientist needs to analyze employment data. The dataset contains approximately 10 million observations on people across 10 different features. During the preliminary analysis, the Data…
The correct answer is B. Numerical value binning D. Logarithmic transformation. Right-skewed distributions (long tail to the right) are common in income and age data. Two effective transformations are: (D) Logarithmic transformation - applying log(x) compresses the long right tail and pulls the distribution closer to normal, making it well-suited for…
Question
A Data Scientist needs to analyze employment data. The dataset contains approximately 10 million observations on people across 10 different features. During the preliminary analysis, the Data Scientist notices that income and age distributions are not normal. While income levels shows a right skew as expected, with fewer individuals having a higher income, the age distribution also shows a right skew, with fewer older individuals participating in the workforce. Which feature transformations can the Data Scientist apply to fix the incorrectly skewed data? (Choose two.)
Options
- ACross-validation
- BNumerical value binning
- CHigh-degree polynomial transformation
- DLogarithmic transformation
- EOne hot encoding
How the community answered
(28 responses)- A11% (3)
- B79% (22)
- C4% (1)
- E7% (2)
Explanation
Right-skewed distributions (long tail to the right) are common in income and age data. Two effective transformations are: (D) Logarithmic transformation - applying log(x) compresses the long right tail and pulls the distribution closer to normal, making it well-suited for positively skewed continuous features like income. (B) Numerical value binning - grouping continuous values into discrete buckets (e.g., age ranges: 20–30, 30–40) reduces the impact of extreme values and removes skew by treating data ordinally. The other options are incorrect: cross-validation (A) is a model evaluation technique, not a feature transformation; high-degree polynomial transformation (C) can amplify skewness; and one-hot encoding (E) is used for nominal categorical variables, not for fixing skewed numerical distributions.
Topics
Community Discussion
No community discussion yet for this question.