IKM Data Analysis and Business Intelligence 3

Question 1

In predictive analytics, what is 'overfitting' in a machine learning model?

Accepted Answer

When a model learns the training data too precisely and performs poorly on new data

Answer

Using too few training records, causing high bias in predictions

Answer

Applying the same model to multiple unrelated business problems

Answer

Normalizing input features to the same numeric scale before training

Question 2

What is a 'Slowly Changing Dimension' (SCD) in data warehousing?

Accepted Answer

A dimension whose attribute values change over time and must be managed historically

Answer

A fact table column whose values change only on a monthly aggregation schedule

Answer

A dimension table with fewer than 100 distinct rows

Answer

A column that stores timestamp data for ETL auditing purposes

Question 3

Which measure of central tendency is LEAST affected by extreme outlier values in a dataset?

Accepted Answer

Median

Answer

Mean

Answer

Mode

Answer

Weighted average

Question 4

In SQL, what is the difference between WHERE and HAVING clauses?

Accepted Answer

WHERE filters rows before aggregation; HAVING filters groups after aggregation

Answer

WHERE applies to text columns only; HAVING applies to numeric columns only

Answer

WHERE is used with joins; HAVING is used with subqueries

Answer

WHERE executes after GROUP BY; HAVING executes before GROUP BY

Question 5

A business analyst creates a report showing that ice cream sales and drowning incidents both rise in summer. What analytical error does this illustrate?

Accepted Answer

Confusing correlation with causation

Answer

Confirmation bias

Answer

Survivorship bias

Answer

Sampling bias

Question 6

What is the primary advantage of using a 'star schema' over a fully normalized schema in a data warehouse?

Accepted Answer

Star schemas simplify queries and improve read performance for analytical workloads

Answer

Star schemas eliminate all data redundancy, reducing storage costs significantly

Answer

Star schemas enforce strict referential integrity through cascading foreign keys

Answer

Star schemas allow real-time transactional writes with minimal locking

Question 7

Which data visualization principle does the concept of 'data-ink ratio' (Edward Tufte) promote?

Accepted Answer

Maximizing the proportion of ink used to convey actual data versus decorative elements

Answer

Using bright colors to highlight the most important data points

Answer

Ensuring every chart includes a legend and title for accessibility

Answer

Using 3D effects to make charts more visually engaging for executives