MLS-C01 · Question #336
An online retailer collects the following data on customer orders: demographics, behaviors, location, shipment progress, and delivery time. A data scientist joins all the collected datasets. The…
The correct answer is B. K-means D. Principal component analysis (PCA). To identify groups of customers from a high-dimensional dataset for a marketing campaign, dimensionality reduction using PCA should precede clustering with K-means to improve efficiency and effectiveness.
Question
An online retailer collects the following data on customer orders: demographics, behaviors, location, shipment progress, and delivery time. A data scientist joins all the collected datasets. The result is a single dataset that includes 980 variables. The data scientist must develop a machine learning (ML) model to identify groups of customers who are likely to respond to a marketing campaign. Which combination of algorithms should the data scientist use to meet this requirement? (Choose two.)
Options
- ALatent Dirichlet Allocation (LDA)
- BK-means
- CSemantic segmentation
- DPrincipal component analysis (PCA)
- EFactorization machines (FM)
How the community answered
(57 responses)- A21% (12)
- B61% (35)
- C12% (7)
- E5% (3)
Why each option
To identify groups of customers from a high-dimensional dataset for a marketing campaign, dimensionality reduction using PCA should precede clustering with K-means to improve efficiency and effectiveness.
Latent Dirichlet Allocation (LDA) is primarily used for topic modeling in natural language processing (NLP) and is not typically applied to tabular customer data for grouping.
K-means is a popular unsupervised clustering algorithm suitable for identifying distinct 'groups of customers' based on their characteristics, directly addressing the core requirement of the problem.
Semantic segmentation is a deep learning technique used in computer vision for pixel-level classification in images and is irrelevant for identifying customer groups from tabular data.
Principal Component Analysis (PCA) is a dimensionality reduction technique that is highly beneficial when dealing with a high number of variables (like 980), as it transforms correlated features into a smaller set of independent components, improving the efficiency and effectiveness of subsequent clustering algorithms like K-means.
Factorization machines (FM) are primarily used for supervised learning tasks like recommendations and predicting ratings by modeling feature interactions, rather than for unsupervised customer grouping.
Concept tested: Dimensionality reduction (PCA) and clustering (K-means)
Source: https://docs.aws.amazon.com/sagemaker/latest/dg/pca.html
Topics
Community Discussion
No community discussion yet for this question.