Back to Subject

Unit 14: Futuristic World of Data Analytics - Practice Quiz

DECAP145 60 Questions
0 Correct 0 Wrong 60 Left
0/60

1 What does the term Big Data primarily refer to?

Introduction to Big data and Analysis Techniques Easy
A. Extremely large and complex datasets that traditional tools struggle to process
B. A type of computer monitor
C. A single small spreadsheet
D. A programming language

2 Which of the following is one of the classic 3 V's used to describe Big Data?

Introduction to Big data and Analysis Techniques Easy
A. Version
B. Vector
C. Vault
D. Volume

3 In the context of Big Data, what does Velocity mean?

Introduction to Big data and Analysis Techniques Easy
A. The color of the data charts
B. The speed at which data is generated and processed
C. The number of data errors
D. The total size of stored data

4 Which V of Big Data refers to the different types and formats of data such as text, images, and video?

Introduction to Big data and Analysis Techniques Easy
A. Velocity
B. Variety
C. Volume
D. Veracity

5 In a dataset, a variable is best described as:

Elements, Variables, and Data categorization Levels of Measurement Easy
A. A type of hardware
B. A characteristic that can take different values
C. The name of the file
D. A fixed unchanging number

6 Which of the following is an example of a categorical (qualitative) variable?

Elements, Variables, and Data categorization Levels of Measurement Easy
A. Annual income
B. Eye color
C. Height in centimeters
D. Weight in kilograms

7 Which level of measurement only classifies data into named categories with no natural order?

Elements, Variables, and Data categorization Levels of Measurement Easy
A. Ratio
B. Nominal
C. Interval
D. Ordinal

8 A ranking such as Small, Medium, Large is an example of which level of measurement?

Elements, Variables, and Data categorization Levels of Measurement Easy
A. Interval
B. Nominal
C. Ratio
D. Ordinal

9 Which level of measurement has a true zero point, allowing meaningful ratios like 'twice as much'?

Elements, Variables, and Data categorization Levels of Measurement Easy
A. Ordinal
B. Interval
C. Nominal
D. Ratio

10 Temperature measured in Celsius, where degrees does not mean 'no temperature', is an example of which level of measurement?

Elements, Variables, and Data categorization Levels of Measurement Easy
A. Ratio
B. Nominal
C. Interval
D. Ordinal

11 In statistics, the individual objects or people on which data is collected are called:

Elements, Variables, and Data categorization Levels of Measurement Easy
A. Labels
B. Formulas
C. Charts
D. Elements

12 What is the main purpose of an index in a database?

Data management and indexing Easy
A. To change data types
B. To reduce the screen brightness
C. To speed up the retrieval of data
D. To delete records automatically

13 Which of the following best describes data management?

Data management and indexing Easy
A. Printing documents on paper
B. Writing marketing emails
C. Designing computer hardware
D. The practice of collecting, storing, and maintaining data

14 A key that uniquely identifies each record in a database table is called a:

Data management and indexing Easy
A. Primary key
B. Foreign gate
C. Data slot
D. Index page

15 What is a possible downside of adding too many indexes to a database table?

Data management and indexing Easy
A. It permanently deletes data
B. It disables all queries
C. It can slow down insert and update operations while consuming extra storage space to maintain the index structures
D. It changes the primary key

16 Which type of statistical learning uses labeled data to make predictions?

Introduction to statistical learning Easy
A. Random guessing
B. Supervised learning
C. Manual sorting
D. Unsupervised learning

17 Unsupervised learning is mainly used to:

Introduction to statistical learning Easy
A. Predict labeled outcomes
B. Find hidden patterns or groupings in unlabeled data
C. Store passwords securely
D. Print reports

18 Which of the following is a common supervised learning task?

Introduction to statistical learning Easy
A. Encrypting
B. Formatting
C. Classification
D. Clustering

19 Which programming language is widely known for data analysis using libraries like Pandas and NumPy?

Overview of various tools used for data analysis Easy
A. CSS
B. SQLite backup
C. Python
D. HTML

20 Which of the following tools is commonly used for creating interactive data visualizations and dashboards?

Overview of various tools used for data analysis Easy
A. Calculator
B. Tableau
C. Paint
D. Notepad

21 A retail company collects streaming clickstream logs, product images, and customer reviews. Which characteristic of Big Data is most directly demonstrated by this mix of formats?

Introduction to Big data and Analysis Techniques Medium
A. Variety
B. Velocity
C. Value
D. Veracity

22 A fraud detection system must process credit card transactions within milliseconds of them occurring. Which 'V' of Big Data is the primary challenge here?

Introduction to Big data and Analysis Techniques Medium
A. Veracity
B. Volume
C. Variety
D. Velocity

23 An analyst wants to answer the question "Why did sales drop last quarter?" Which type of analytics best fits this need?

Introduction to Big data and Analysis Techniques Medium
A. Prescriptive analytics
B. Predictive analytics
C. Diagnostic analytics
D. Descriptive analytics

24 A logistics firm uses a model that not only forecasts demand but also recommends optimal delivery routes to act on. This is an example of which analytics type?

Introduction to Big data and Analysis Techniques Medium
A. Diagnostic analytics
B. Descriptive analytics
C. Prescriptive analytics
D. Predictive analytics

25 In a dataset table, each row represents a surveyed customer and each column stores an attribute such as age or income. What do the rows correspond to?

Elements, Variables, and Data categorization Levels of Measurement Medium
A. Data types
B. Variables
C. Measurement levels
D. Elements (observations)

26 A variable stores 'number of children in a household' as . How is this variable best classified?

Elements, Variables, and Data categorization Levels of Measurement Medium
A. Ordinal categorical variable
B. Nominal categorical variable
C. Discrete quantitative variable
D. Continuous quantitative variable

27 Temperature measured in degrees Celsius has meaningful differences but its zero point does not represent 'no temperature'. Which level of measurement does it belong to?

Elements, Variables, and Data categorization Levels of Measurement Medium
A. Ratio
B. Interval
C. Ordinal
D. Nominal

28 A survey records satisfaction as Poor, Average, Good, Excellent. Which statement about this variable is correct?

Elements, Variables, and Data categorization Levels of Measurement Medium
A. It is nominal because the categories are just labels
B. It is ordinal because categories have a meaningful order but unequal gaps
C. It is interval because the gaps between ratings are equal
D. It is ratio because it has a true zero point

29 Which operation is valid on ratio-level data but NOT on interval-level data?

Elements, Variables, and Data categorization Levels of Measurement Medium
A. Computing meaningful ratios such as 'twice as much'
B. Calculating the arithmetic mean
C. Ranking values from lowest to highest
D. Measuring equal differences between values

30 Assigning the numbers and in a dataset is an example of encoding which type of variable?

Elements, Variables, and Data categorization Levels of Measurement Medium
A. Interval
B. Ordinal
C. Ratio
D. Nominal

31 A query on a large table filters frequently by the customer_id column but runs slowly. Which technique would most directly improve retrieval speed for this column?

Data management and indexing Medium
A. Creating an index on customer_id
B. Deleting duplicate rows
C. Adding more columns to the table
D. Converting the column to text type

32 Which is a common trade-off of adding many indexes to a database table?

Data management and indexing Medium
A. Reduced storage but slower queries
B. Faster reads but slower inserts and updates
C. Improved reads and writes with no cost
D. Faster inserts but slower reads

33 During data cleaning, an analyst finds the same customer stored as "John Smith" and "john smith". This is an example of which data quality issue?

Data management and indexing Medium
A. Inconsistent data
B. Missing data
C. Outlier data
D. Duplicate index

34 A primary key in a relational table must satisfy which requirement?

Data management and indexing Medium
A. It must uniquely identify each row and not be null
B. It must be the first column in the table
C. It must be a numeric value only
D. It may repeat as long as values are ordered

35 A model is trained on labeled housing data to predict a home's selling price from its features. This is an example of which learning type?

Introduction to statistical learning Medium
A. Unsupervised clustering
B. Supervised classification
C. Supervised regression
D. Reinforcement learning

36 An algorithm groups customers into segments based on purchasing behavior without any predefined labels. Which category does this task fall under?

Introduction to statistical learning Medium
A. Classification
B. Unsupervised learning
C. Supervised learning
D. Regression

37 A model performs very well on training data but poorly on new test data. This situation is best described as:

Introduction to statistical learning Medium
A. Underfitting
B. Regularization
C. Overfitting
D. Cross-validation

38 Which task correctly matches a classification problem?

Introduction to statistical learning Medium
A. Predicting tomorrow's exact temperature
B. Predicting whether an email is spam or not spam
C. Forecasting a company's revenue in dollars
D. Estimating a person's monthly income

39 Why is a dataset typically split into training and test sets in statistical learning?

Introduction to statistical learning Medium
A. To evaluate how well the model generalizes to unseen data
B. To convert categorical variables into numbers
C. To remove all outliers from the dataset
D. To increase the total amount of available data

40 A data scientist needs a programming language with rich libraries such as pandas and scikit-learn for end-to-end machine learning. Which tool is most appropriate?

Overview of various tools used for data analysis Medium
A. MySQL
B. Python
C. Microsoft Excel
D. Tableau

41 A streaming platform ingests sensor readings arriving at millions of events per second, where a delay of even a few seconds makes the data useless for its real-time dashboards. Which of the 5 Vs of Big Data is the primary constraint being described here?

Introduction to Big data and Analysis Techniques Hard
A. Volume
B. Variety
C. Veracity
D. Velocity

42 An analyst uses historical sales data to build a model that recommends what actions to take to maximize next quarter's revenue, factoring in constraints. Which category of analytics does this represent?

Introduction to Big data and Analysis Techniques Hard
A. Prescriptive analytics
B. Descriptive analytics
C. Diagnostic analytics
D. Predictive analytics

43 In the Hadoop ecosystem, why does the MapReduce paradigm move computation to the data node rather than moving data to the computation?

Introduction to Big data and Analysis Techniques Hard
A. It eliminates the need for a shuffle-and-sort phase between stages
B. Network bandwidth is the bottleneck, so processing data locally minimizes costly data transfer
C. It guarantees the data is never replicated across the cluster
D. CPU cores on data nodes are faster than on compute nodes

44 A researcher records water temperature in degrees Celsius. A colleague claims " is twice as hot as ." Why is this statement statistically invalid?

Elements, Variables, and Data categorization Levels of Measurement Hard
A. Celsius is a nominal scale, so no arithmetic is allowed
B. Celsius is an interval scale with an arbitrary zero, so ratios are meaningless
C. Celsius is a ratio scale, so the statement is actually correct
D. Celsius is an ordinal scale, so only ranking is valid

45 A dataset codes customer satisfaction as Poor, Fair, Good, Excellent. Which statistical operation is inappropriate for this variable?

Elements, Variables, and Data categorization Levels of Measurement Hard
A. Ranking respondents by satisfaction
B. Determining the median category
C. Reporting the mode
D. Computing the arithmetic mean of the codes

46 Which of the following variables is best classified as discrete quantitative, not continuous?

Elements, Variables, and Data categorization Levels of Measurement Hard
A. The time taken to process an order
B. The temperature of a warehouse
C. The exact weight of a shipment in kilograms
D. The number of defective items in a shipment

47 In an experiment, ZIP codes are stored as integers. An algorithm computes their average and uses it as a feature. What is the fundamental error?

Elements, Variables, and Data categorization Levels of Measurement Hard
A. ZIP codes are ordinal, so only the median should be used
B. ZIP codes are ratio data, so averaging is valid but rounding is needed
C. ZIP codes are nominal labels, so arithmetic on them is meaningless
D. ZIP codes are interval data, so the mean is acceptable

48 A database table with millions of rows has a B-tree index on a column with only two distinct values (Y/N), roughly evenly split. Why might the query optimizer ignore this index for a filter like WHERE flag = 'Y'?

Data management and indexing Hard
A. Indexes only work on primary key columns
B. Low cardinality means a full table scan is often cheaper than many random index lookups
C. The index is automatically dropped when cardinality is below three
D. B-tree indexes cannot store character data

49 Adding many indexes to a heavily transactional (OLTP) table can degrade performance. What is the main reason?

Data management and indexing Hard
A. Indexes force all queries to run as full table scans
B. Indexes double the storage required for every read operation
C. Each INSERT, UPDATE, and DELETE must also maintain every index, increasing write cost
D. Indexes convert the table into a read-only structure

50 A composite index is created on columns (last_name, first_name). Which query can use this index efficiently for its search?

Data management and indexing Hard
A. WHERE last_name = 'Rao'
B. WHERE city = 'Pune'
C. WHERE first_name = 'Asha'
D. WHERE first_name = 'Asha' AND city = 'Pune'

51 In the ACID properties of transactional data management, which property is directly threatened when two concurrent transactions read and write the same row without proper locking?

Data management and indexing Hard
A. Isolation
B. Atomicity
C. Consistency
D. Durability

52 A model achieves accuracy on training data but only on unseen test data. This gap is the classic symptom of which problem, and what is the appropriate remedy?

Introduction to statistical learning Hard
A. Overfitting; apply regularization or reduce model complexity
B. High bias; remove regularization entirely
C. Data leakage; increase the training set size only
D. Underfitting; increase model complexity further

53 In the bias-variance tradeoff, what happens to a model's total expected error as flexibility (complexity) increases from very low to very high?

Introduction to statistical learning Hard
A. Bias decreases while variance increases, so error follows a U-shape
B. Bias increases while variance decreases, so error keeps falling
C. Both bias and variance increase continuously
D. Both bias and variance decrease continuously

54 A dataset for fraud detection has legitimate and fraudulent transactions. A model predicts "legitimate" for every case. Why is its accuracy misleading?

Introduction to statistical learning Hard
A. Accuracy ignores the minority class; metrics like recall or F1 are needed
B. The model is actually underfitting the majority class
C. Accuracy is being computed on the wrong test set
D. Accuracy should be replaced by training error

55 Which statement correctly distinguishes supervised from unsupervised learning?

Introduction to statistical learning Hard
A. Both require labeled data but differ in the loss function used
B. Supervised learning uses labeled outputs to guide training; unsupervised finds structure in unlabeled data
C. Supervised learning is only for clustering; unsupervised is only for regression
D. Supervised learning requires no target variable; unsupervised always predicts a label

56 Why does -fold cross-validation generally give a more reliable estimate of test error than a single train-test split?

Introduction to statistical learning Hard
A. It always produces lower error because it trains on all data at once
B. Every observation is used for both training and validation across folds, reducing estimate variance
C. It removes the need for a separate test set by eliminating bias entirely
D. It guarantees the model will not overfit the training data

57 A team must process a dataset that exceeds a single machine's RAM, using in-memory transformations across a cluster. Which tool is most appropriate?

Overview of various tools used for data analysis Hard
A. Apache Spark
B. A standalone MySQL instance
C. A single-node pandas script
D. Microsoft Excel

58 When deciding between R and Python for a project combining heavy statistical modeling with a production web-service deployment, which consideration best justifies choosing Python?

Overview of various tools used for data analysis Hard
A. R cannot perform any statistical analysis
B. Python has no libraries for data manipulation
C. R is unable to produce any data visualizations
D. Python integrates more easily into general-purpose software and production pipelines

59 In SQL, why can WHERE filter on a column but HAVING is required to filter on the result of COUNT(*) in a grouped query?

Overview of various tools used for data analysis Hard
A. HAVING runs before WHERE, so it sees raw rows first
B. WHERE filters rows before grouping; aggregate results exist only after grouping, which HAVING handles
C. COUNT(*) is only valid inside a WHERE clause
D. WHERE cannot reference any column in the table

60 A business analyst needs interactive, self-service dashboards for non-technical executives, refreshed from a data warehouse. Which category of tool is the best fit?

Overview of various tools used for data analysis Hard
A. A low-level command-line ETL script
B. A unit-testing framework
C. A raw NoSQL key-value store
D. A business intelligence tool such as Tableau or Power BI
Previous MCQ
Unit 13