Q36. You are working with two variables:

is_local_guide: 1 if the reviewer is a Local Guide, missing otherwise
review_text: text of the review, missing if the user did not leave a comment

Which of the following correctly handles missing values?

Question

Q36. You are working with two variables:

is_local_guide: 1 if the reviewer is a Local Guide, missing otherwise
review_text: text of the review, missing if the user did not leave a comment

Which of the following correctly handles missing values?

BlackTom AI · Accepted Answer

Binary indicator variables and unstructured text must be handled differently: is_local_guide is a binary (1 = Local Guide, missing otherwise) so missing values semantically mean "not a Local Guide" and should be set to 0; review_text is unstructured text, so representing a missing comment as numeric 0 is inappropriate — leave it missing, use an empty string, or encode a separate "has_review" flag.

1) Relevant knowledge points and concepts
- Variable types: categorical/binary vs. unstructured text require different imputation strategies.
- Meaning of missingness: if the data definition states missing implies absence, then impute the absence value (0 for a binary indicator).
- Imputation appropriateness: numeric imputation (mean, 0) only makes sense for numeric variables; text requires different handling (keep missing, empty string, or a separate indicator).
- Bias and data loss: dropping rows with missing values can remove valid and informative records and introduce bias.

2) Step-by-step reasoning
- Inspect is_local_guide: defined as 1 if Local Guide, missing otherwise. Therefore missing = not a Local Guide. Logical imputation: missing -> 0.
- Inspect review_text: this is free text; missing means the user did not leave a comment. Replacing text with 0 would mix data types and falsely imply a numeric value. Better options: keep as missing/NaN, replace with empty string "", or create an additional binary feature has_review = 0/1.

3) Intermediate mapping example
- is_local_guide: {1 -> 1, missing -> 0}
- review_text: {text -> text, missing -> NaN or "" (and optionally has_review = 0)}

4) Why the correct answer is correct
Replacing missing values with 0 in is_local_guide preserves the intended semantics (absence = 0). Not replacing review_text with 0 preserves data type and avoids introducing invalid numeric values; missing text should be encoded differently.

5) Why other options are incorrect (brief)
- Replace missing with 0 in both: wrong because 0 is not a valid representation for text.
- Replace missing with mean in both: mean is meaningless for text; for binary it would introduce fractional/ambiguous values and distort interpretation.
- Drop all rows with any missing: discards valid observations and biases the dataset.
- Treat missing values the same across variables: ignores differing types and meanings of missingness.

Conclusion: The correct strategy is to replace missing values with 0 in is_local_guide, but not in review_text — matching the provided answer.

Q36. You are working with two variables: is_local_guide: 1 if the reviewer is a Local Guide, missing otherwise review_text: text of the review, missing if the user did not leave a comment Which of the following correctly handles missing values?单项选择题

类似问题

Which of the following data validation techniques can be used to handle missing values?

Which of the following is NOT a suggested way to address empty fields?

Coding Question: Which method would you use in Python to drop missing values in a DataFrame?

You own a bond that pays $64 in interest annually. The face value is $1,000 and the current market price is $1,021.61. The bond matures in 11 years. What is the yield to maturity?

Which of the following statements is CORRECT?

更多留学生实用工具

智能学习助手

风格化写作助手

论文查重助手

文献引用助手

课堂转译助手

课堂笔记助手

Quiz搜索助手

学校历年真题

智能刷题助手

智能匹配练习

希望你的学习变得更简单