← All IKM Flashcard Decks

Data Analysis and Business Intelligence Flashcards

7 cards from real IKM practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.

Read the first 7 Data Analysis and Business Intelligence flashcards as text
  1. In predictive analytics, what is 'overfitting' in a machine learning model?

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

    Overfitting occurs when a model memorizes noise and detail in training data to such a degree that it fails to generalize to new, unseen data, resulting in poor predictive performance in production.

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

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

    A Slowly Changing Dimension (SCD) is a dimension whose attributes change slowly over time (e.g., a customer's address), requiring strategies like Type 1, 2, or 3 to track or overwrite historical values.

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

    Answer: Median

    The median is the middle value when data is sorted, making it resistant to outliers because extreme values do not shift it significantly, unlike the mean which is pulled toward outliers.

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

    Answer: WHERE filters rows before aggregation; HAVING filters groups after aggregation

    WHERE filters individual rows before any grouping or aggregation occurs, while HAVING filters the resulting groups after a GROUP BY clause and aggregate functions have been applied.

  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?

    Answer: Confusing correlation with causation

    This illustrates confusing correlation with causation — both variables are driven by a common third factor (hot weather), but ice cream sales do not cause drownings; they are merely correlated.

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

    Answer: Star schemas simplify queries and improve read performance for analytical workloads

    Star schemas denormalize dimension data into flat tables around a central fact table, which reduces the number of joins required for analytical queries and improves query performance for BI tools.

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

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

    Tufte's data-ink ratio principle states that every element of a visualization should serve a purpose in conveying data — excess gridlines, borders, and decorative elements ('chartjunk') should be removed.