CCNA Exploratory Data Analysis Questions

Using pandas .isnull().sum() in a SageMaker notebook is a standard approach to count missing values per column. Option A is wrong because S3 Select is for filtering S3 objects, not for data analysis. Option B is wrong because QuickSight is for visualization but not for programmatic missing value analysis.

Option D is wrong because Athena requires SQL and is less direct for EDA. Option E is wrong because Glue Crawler discovers schema, not missing values.

Option C is correct because TF-IDF downweights common words. Option A is wrong because bag-of-words does not weight. Option B is wrong because removing stopwords helps but does not adjust for frequency beyond that.

Option D is wrong because word2vec focuses on context, not frequency weighting.

Option A is correct because stratified sampling ensures that the minority class is adequately represented in the sample. Option B is wrong because random sampling may miss the rare class entirely. Option C is wrong because SMOTE is a data augmentation technique for training, not for EDA.

Option D is wrong because removing the minority class would prevent analyzing the target.

Option A is correct because permutation importance is model-agnostic and measures feature importance by shuffling. Option B is wrong because SHAP values are model-specific. Option C is wrong because L1 regularization is model-specific.

Option D is wrong because correlation is bivariate.

Option A is correct because zero-inflated models handle excess zeros. Option B is wrong because linear regression assumes normal distribution. Option C is wrong because Poisson regression is for count data without excess zeros.

Option D is wrong because logistic regression is for binary outcomes.

Options A and D are correct. Count plot and value_counts show class frequencies. Option B is wrong because correlation matrix shows relationships between features.

Option C is wrong because scatter plot shows relationship between two numeric variables. Option E is wrong because histogram shows distribution of a continuous variable.

Option A is correct because correlation-based feature selection removes redundant features, and target encoding handles high-cardinality categoricals without expanding dimensions. Option B is wrong because PCA reduces interpretability and does not handle categoricals. Option C is wrong because chi-squared test is for categorical targets, but dataset has binary target; also one-hot encoding explodes dimensions.

Option D is wrong because mutual information is used for feature selection but does not address high cardinality directly.

Grouping rare categories into an 'other' bucket is a common technique to reduce cardinality. Option A (one-hot encoding) would create too many columns. Option B (label encoding) doesn't reduce uniqueness.

Option D (frequency encoding) replaces categories with frequency but still has 1000 values.

Option D is correct because if total_charges is missing only for tenure=0, it means those customers have not been billed yet, so total_charges should be 0. Imputing with 0 is appropriate. Option A is wrong because dropping rows with missing total_charges would remove all new customers, biasing the dataset.

Option B is wrong because imputing with mean would assign incorrect values to new customers. Option C is wrong because using a model to predict missing values is overkill and may introduce error when the true value is known to be 0.

Option D is correct because a time series line plot with two lines (fraud vs. non-fraud) shows temporal patterns. Option A is wrong because bar chart of counts per day is less effective for two categories. Option B is wrong because scatter plot with 1 million points is overwhelming.

Option C is wrong because box plot shows distribution per time period but not temporal trend.

Option C is correct because Chi-squared test is used for independence between categorical variables. Option A is wrong because ANOVA is for continuous target. Option B is wrong because Pearson correlation is for continuous variables.

Option D is wrong because Kolmogorov-Smirnov test is for distribution comparison.

Option B is correct because the command filters objects larger than 100000 bytes, and the output shows three large files. Option A is wrong because the command does not count all objects; it filters by size. Option C is wrong because the output shows three files, not all files.

Option D is wrong because the command explicitly filters by size, so the output is not all objects.

Options A, B, and D are correct. A: Correlation matrix shows pairwise correlations; high values indicate collinearity. B: Variance Inflation Factor (VIF) quantifies how much a feature is explained by others.

D: Eigenvalues from PCA can indicate multicollinearity if some are near zero. Option C is incorrect because a scatter matrix shows pairwise relationships but not a multicollinearity measure. Option E is incorrect because a heatmap of missing values is unrelated to multicollinearity.

Options B and D are correct. Amazon QuickSight supports geospatial charts and time-series. Amazon SageMaker Data Wrangler can also visualize geospatial data as scatter plots.

Option A is wrong because Amazon Comprehend is for NLP. Option C is wrong because AWS Glue is for ETL. Option E is wrong because Amazon Rekognition is for image/video analysis.

PCA (Principal Component Analysis) is the standard technique for dimensionality reduction by projecting data onto principal components that capture maximum variance. LDA is supervised and aims to separate classes. t-SNE is for visualization. Autoencoders can reduce dimensionality but are more complex.

Factor analysis assumes latent factors.

Option A is correct because log transformation is commonly used to reduce right skewness. Option B is wrong because square transformation increases skewness. Option C is wrong because Min-Max scaling does not change shape.

Option D is wrong because one-hot encoding is for categorical features.

Option C is correct because the Deny statement explicitly denies s3:* actions on the confidential folder unless the request comes from a specific VPC endpoint. If the request is from outside that endpoint or from a different VPC, it will be denied. Option A is wrong because the Deny overrides the Allow.

Option B is wrong because the Deny is conditional. Option D is wrong because the condition restricts access.

Option B is correct because a correlation matrix quantifies linear relationships between numerical features. Option A is wrong because box plots show distribution, not relationships. Option C is wrong because chi-square test is for categorical associations.

Option D is wrong because pair plots visualize scatter plots but not a quantitative measure of multicollinearity.

Options A, B, and C are correct. Scatter plots with a LOESS curve visually reveal non-linearity. Polynomial features allow linear models to capture non-linear relationships.

Decision trees can model non-linear interactions without explicit feature engineering. Option D is wrong because correlation measures only linear relationships. Option E is wrong because logistic regression is for binary outcomes.

Option C is correct because removing one of the highly correlated features reduces multicollinearity. Option A is wrong because PCA is not necessary for just two correlated features. Option B is wrong because standard scaling does not address correlation.

Option D is wrong because adding interaction terms increases multicollinearity.

Using Amazon SageMaker Data Wrangler to generate histograms segmented by class is a straightforward way to visualize feature distributions for each class. Option B (t-tests) may be used later but doesn't provide distribution visualization. Option C (box plots) is a good alternative but not as comprehensive as histograms for distribution shape.

Option D (correlation matrix) does not show class-wise distribution.

Option D is correct because multiple imputation by chained equations (MICE) handles MAR well by modeling each variable with missing values conditional on others. Option A is wrong because mean imputation underestimates variance. Option B is wrong because dropping rows with missing data reduces sample size and can introduce bias.

Option C is wrong because KNN imputation assumes data are MCAR and may not be optimal for MAR.

Option C is correct because the IQR method is robust to skewed distributions and does not assume normality. Option A is wrong because Z-score assumes normality. Option B is wrong because Mahalanobis distance assumes multivariate normality.

Option D is wrong because DBSCAN is computationally expensive on 1 million rows and may not be practical for univariate outlier detection.

Option D is correct because Amazon Athena uses a serverless query engine that scales automatically and charges per query based on data scanned, making it cost-effective for large datasets. Option A is wrong because downloading to SageMaker Studio may incur high data transfer costs and require significant local storage. Option B is wrong because AWS Glue Spark jobs have overhead and are more suited for complex ETL.

Option C is wrong because Amazon EMR requires provisioning clusters and is more expensive for simple statistics.

Option C is correct because target encoding replaces categories with the mean of the target, which is efficient and works well with tree models. Option A is wrong because one-hot encoding creates many columns for high cardinality. Option B is wrong because label encoding imposes ordinality.

Option D is wrong because frequency encoding may not capture predictive information.

Option A is correct because PCA is a dimensionality reduction technique that identifies the directions (principal components) that maximize variance. Option B is wrong because t-SNE is for visualization and does not provide variance contributions. Option C is wrong because LDA is supervised and requires labels.

Option D is wrong because K-means is clustering, not variance analysis.