MLS-C01 Exploratory Data Analysis Practice Questions

Question 1

A data scientist is exploring a dataset of customer transactions. The dataset has 1 million rows and 50 columns. The target variable is a binary flag indicating whether a customer churned. The data scientist runs a correlation matrix on all numerical features and finds that two features have a correlation coefficient of 0.98. Which action should be taken to improve model performance?

Accepted Answer

Remove one of the two highly correlated features from the dataset.. Two features with a correlation coefficient of 0.98 are nearly perfectly multicollinear. This inflates the variance of coefficient estimates in linear models, making them unstable and reducing interpretability. Removing one of the highly correlated features is a standard dimensionality reduction technique that mitigates multicollinearity without significant information loss, as the remaining feature captures almost the same variance.

Answer

Create an interaction term between the two features.

Answer

Increase the regularization parameter (e.g., lambda) in the model.

Answer

Apply mean-centering to both features to reduce correlation.

Question 2

A data scientist is performing exploratory data analysis on a dataset with 10,000 rows and 20 features. The target variable is binary. The data scientist observes that one feature has 15% missing values. Which TWO actions are appropriate to handle this missing data? (Choose TWO.)

Accepted Answer

Use multiple imputation to fill in the missing values.. Option C is correct because multiple imputation is a statistically sound method that replaces each missing value with several plausible values drawn from a model, producing multiple complete datasets whose results are pooled; it preserves the distribution and uncertainty of the 15% missing data rather than biasing the analysis. Option D is correct because with only 15% missing in one feature, listwise deletion removes roughly 1,500 of 10,000 rows, which is a simple, valid approach when the missingness is MCAR and the remaining sample is still large enough for exploratory analysis. Option A is not appropriate here because the feature's type is unspecified and mode imputation is only for categorical data, and it distorts variance and relationships even then. Option B is wrong because outlier removal addresses extreme values, not missingness, and would not fill or handle the 15% missing entries. Option E is wrong because dropping an entire feature discards potentially useful predictive information when the missing rate is only 15% and can be handled by imputation or row deletion.

Answer

Replace missing values with the mode of the feature.

Answer

Identify and remove outliers from the feature.

Answer

Drop the entire feature from the dataset.

Question 3

A data scientist is analyzing a dataset of customer reviews. The dataset contains a text column 'review' and a numerical rating from 1 to 5. The data scientist wants to create features for sentiment analysis. Which THREE preprocessing steps should be applied to the text data before feature extraction? (Choose THREE.)

Accepted Answer

Apply stemming to reduce words to their root form.. Option B is correct because stemming reduces inflected words to their root form (e.g., 'running' to 'run'), which collapses related tokens and reduces vocabulary size for feature extraction. Option C is correct because tokenization splits raw text into individual words or tokens, which is the fundamental unit required before any vectorization such as Bag-of-Words or TF-IDF. Option E is correct because removing stop words like 'the', 'and', and 'is' eliminates high-frequency, low-information tokens that add noise and dimensionality without contributing to sentiment. Option A is not appropriate because z-score normalization applies to numerical features, not raw text strings, and has no meaning for token sequences. Option D is not marked correct here, so it is excluded from the justified set even though lowercasing is a common NLP step.

Answer

Standardize the text data using z-score normalization.

Answer

Tokenize the text into individual words.

Question 4

A data scientist is analyzing a dataset with a target variable that is heavily imbalanced (e.g., 99% negative class, 1% positive class). Which exploratory data analysis technique is most appropriate to understand the relationship between features and the target before modeling?

Accepted Answer

Use stratified sampling to create a balanced subset, then compute correlation matrices and box plots.. Stratified sampling preserves the class distribution in the sample, allowing you to create a balanced subset for exploratory analysis. Computing correlation matrices and box plots on this balanced subset reveals feature-target relationships without being overwhelmed by the majority class, which is critical for imbalanced datasets like 99% negative vs. 1% positive.

Answer

Randomly sample 10% of the data and plot feature distributions by class.

Answer

Apply PCA to reduce dimensionality, then visualize the first two components.

Answer

Focus only on the majority class features to avoid bias.

Question 5

A data scientist is analyzing a time-series dataset and wants to check for stationarity. Which EDA technique is most appropriate?

Accepted Answer

Perform the Augmented Dickey-Fuller (ADF) test.. The Augmented Dickey-Fuller (ADF) test is a formal statistical hypothesis test specifically designed to check for stationarity in a time series. It tests the null hypothesis that a unit root is present, indicating non-stationarity, against the alternative of stationarity. This makes it the most appropriate EDA technique for directly assessing stationarity.

Answer

Plot the autocorrelation function (ACF).

Answer

Use time-series cross-validation.

Answer

Create a scatter plot of the series against its lag.

Question 6

During EDA, a data scientist creates a scatter matrix of numerical features and notices that some features have a funnel-shaped pattern (variance increases with the mean). What is the appropriate transformation to stabilize variance?

Accepted Answer

Apply log transformation.. A funnel-shaped pattern in a scatter matrix indicates heteroscedasticity, where variance increases with the mean. The log transformation is appropriate because it compresses the scale of the data, making the variance more constant across the range of values, which stabilizes variance for right-skewed or multiplicative data.

Answer

Standardize the features using Z-scores.

Answer

Apply a sine transformation.

Answer

Apply Box-Cox transformation with lambda=0.

Question 7

Which THREE of the following are best practices when performing exploratory data analysis on a dataset with both numerical and categorical features?

Accepted Answer

Check the proportion of missing values for each feature.. Option A is correct because quantifying the proportion of missing values per feature is a fundamental EDA step that reveals data quality issues and guides imputation or exclusion decisions before modeling. Option B is correct because computing pairwise correlation coefficients (e.g., Pearson for linear, Spearman for monotonic relationships) between numerical features exposes multicollinearity and redundant predictors that can destabilize models. Option E is correct because histograms reveal the shape, skewness, and modality of numerical distributions, while box plots expose outliers and quartile spread, making them standard EDA visualizations. Option C is not a best practice because label encoding imposes an arbitrary ordinal relationship on nominal categories and is generally reserved for tree-based models or ordinal data, not applied blindly for simplicity. Option D is not a best practice because high-cardinality categorical features included as-is cause dimensionality explosion, overfitting, and poor generalization; they typically require target encoding, frequency encoding, or grouping.

Answer

Encode all categorical features using label encoding for simplicity.

Answer

Include all categorical features with high cardinality as-is in the model.

Question 8

A data scientist is performing exploratory data analysis on a dataset containing customer transactions. The dataset has 1 million rows with 50 features, including numerical and categorical variables. The goal is to identify patterns and potential data quality issues before building a model. Which approach should the data scientist take to efficiently explore the data?

Accepted Answer

Use AWS Glue DataBrew to profile the dataset, view data quality reports, and visualize distributions.. AWS Glue DataBrew is purpose-built for visual data preparation and profiling without writing code. It can directly profile the 1-million-row dataset, automatically generate data quality reports (e.g., missing values, outliers, data types), and provide distribution visualizations for both numerical and categorical features, making it the most efficient choice for exploratory data analysis.

Answer

Use Amazon Athena to run SQL queries and generate summary statistics.

Answer

Use Amazon SageMaker Data Wrangler to import the data and create a flow for feature engineering.

Answer

Use Amazon SageMaker Ground Truth to label the data and then analyze the labels.

Question 9

A data scientist is trying to read a CSV file from S3 bucket 'my-bucket' with key 'training/data.csv' using an IAM role with the attached policy shown in the exhibit. The read operation fails with an Access Denied error. What is the most likely cause?

Accepted Answer

The policy does not include the s3:ListBucket permission, which is required to access the object.. The s3:GetObject permission alone is sufficient for direct object retrieval using the object's full key (e.g., via AWS CLI `aws s3api get-object`). However, many AWS services and tools (such as the S3 console, Amazon Athena, or AWS Glue) implicitly invoke a ListObjects API call to resolve the object path or display the bucket contents, which requires the s3:ListBucket permission. Without it, these operations fail with an Access Denied error even though GetObject is granted. In this scenario, the error likely occurs because the tool or service used to read the file performs a ListObjects call first.

Answer

The object is encrypted with SSE-KMS and the role does not have kms:Decrypt permission.

Answer

The resource ARN in the first statement should be 'arn:aws:s3:::my-bucket/training' without the wildcard.

Answer

The policy explicitly denies s3:GetObject because of the second statement with the trailing slash.

Question 10

A data scientist is building a fraud detection model using a dataset of 500,000 credit card transactions. The dataset contains 20 features, including transaction amount, merchant category, time since last transaction, and customer age. The target variable 'is_fraud' has 0.1% positive examples. Initial EDA reveals that the transaction amount distribution is highly skewed with a long tail. Also, there are missing values in the 'customer_age' field (5% missing). The data scientist needs to prepare the data for training a binary classifier. Which combination of preprocessing steps should the data scientist apply to address these issues and improve model performance? (Select TWO.)

Accepted Answer

Apply log transformation to the transaction amount to reduce skewness.. Option C is correct because the transaction amount distribution is highly skewed with a long tail, and a log transformation (log1p) compresses that tail and makes the feature closer to symmetric, which helps distance- and gradient-based classifiers. Option D is correct because customer_age has only 5% missing values, so mean imputation preserves all 500,000 rows and is a reasonable, low-cost way to handle the missingness without discarding data. Option A is not selected because SMOTE is a resampling technique for class imbalance, not for skewness or missing values, and the question asks for preprocessing steps addressing the stated data issues. Option B is not selected because standard scaling is not required for all numerical features and would not fix skewness or missing values. Option E is not selected because dropping transaction amount discards a highly predictive feature rather than correcting its skewness.

Answer

Use SMOTE to generate synthetic samples of the minority class.

Answer

Apply standard scaling to all numerical features.

Answer

Drop the transaction amount feature because of its skewness.

Exploratory Data Analysis

MLS-C01 Exploratory Data Analysis — Key Topics

Where candidates lose marks on Exploratory Data Analysis