A pipeline pulls customer addresses from three source systems that each store state names differently: ‘California’, ‘CA’, and ‘06’. Which Transform operation should be applied?

Question

BlackTom AI · Accepted Answer

Normalization/standardization — map all variants to a single consistent representation is the appropriate Transform operation.

Relevant knowledge points and concepts:
- Data normalization/standardization (also called canonicalization) converts heterogeneous representations of the same entity into a single agreed-upon format (controlled vocabulary).  
- Common address fields (state) are often stored as full names, postal abbreviations, or numeric codes (e.g., FIPS). A mapping/lookup table or transformation function is used to reconcile these.  
- Other transform categories (type casting, feature engineering, deduplication) serve different purposes and do not reconcile semantic variants.

Step-by-step reasoning and intermediate mapping:
1. Identify input variants for the state field: "California", "CA", "06". Recognize these represent the same state (06 is the FIPS code for California).  
2. Choose a target canonical representation (example: USPS two-letter code "CA" or full name "California"). Suppose we choose "CA".  
3. Build a mapping table (lookup):  
   - "California" -> "CA"  
   - "CA" -> "CA"  
   - "06" -> "CA" (if incoming is numeric string, strip leading zeros or map "06" explicitly)  
4. Apply transformation: for each record, normalize casing/whitespace, then lookup the canonical value and replace the stored value with "CA". Records not found in the lookup are flagged for review.

Why the correct answer is correct:
- The problem requires unifying different textual and coded representations into one consistent form. That is the precise goal of normalization/standardization. A lookup or mapping routine will ensure all three source formats are represented identically downstream.

Why other options are incorrect:
- Data type casting — converting string to integer would not reconcile "California" or "CA" to the same form; it only changes datatype.  
- Feature engineering — creating a derived column is for predictive features, not for harmonizing existing semantic values.  
- Deduplication — removes duplicate records but does not map different representations of the same value to a single representation.

Conclusion: The correct Transform operation is "Normalization/standardization — map all variants to a single consistent representation," which matches the provided answer.

A pipeline pulls customer addresses from three source systems that each store state names differently: ‘California’, ‘CA’, and ‘06’. Which Transform operation should be applied?单项选择题

类似问题

You are preprocessing a dataset for a machine learning model and decide to apply normalization. Which of the following statements about normalization are correct? (Select all that apply.)

Normalization is especially important before non-linearity such as sigmoid or tanh

Is the following statement true or false? In the products of methylation-hydrolysis, every -OH group corresponds to the position of a glycosidic bond in the starting polysaccharide.

Which of the follwoing structures represents amylopectin?

Which of the following is true about an aldopentose?I. It is a monosaccharide.II. It contains a CHO groupIII. It is a disaccharide.IV. It is an oligosaccharide.

For stock A, we have 𝛽 𝑖 = 0.70. Suppose the expected market risk premium next year is 9% and the risk-free rate is 3%. What is the expected return of this stock based on the CAPM? (Please answer in % and round to 2 decimal places. If the answer is 8.057%, then in the box, write 8.06)

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