NEW
Font size
WorksheetsChapter 1 – Data Science Worksheet (Grade 12)
Total questions: 79
Worksheet time: 40mins
Who first proposed the term Data Science?
William S. Cleveland
Chien-Fu Jeff Wu
DJ Patil
John Tukey
What was William S. Cleveland’s key contribution to Data Science?
Popularized the concept of Big Data
Proposed Data Science as an independent discipline
Developed the R programming language
Founded Kaggle
The growth of Data Science in the 21st century is mainly due to which factor?
Decline of the Internet
Slow progress in software
Big data and high computational power
Decreased demand for analytics
Who helped popularize the term Data Scientist?
LinkedIn and Facebook
IBM
Microsoft
How does Data Science mainly differ from Data Analysis?
It focuses only on statistics
It covers the entire process from data collection to deployment
It does not require programming
It ignores understanding data
Which factor makes Data Science important for businesses?
Reduced productivity
Increased ability to make decisions based on data
Lower reliability
Technology limitations
In healthcare, Data Science can help with which task?
Analyzing medical images and predicting diseases
Reducing processing speed
Increasing treatment costs
Removing data
An application of Data Science in finance is:
Predicting market trends and credit risk
Designing website interfaces
Recording transactions
Deleting financial history
In education, Data Science can help with:
Randomly ranking students
Analyzing learning outcomes and optimizing curricula
Reducing the amount of data
Limiting the number of students
In agriculture, Data Science helps mainly by:
Reducing productivity
Predicting weather and optimizing crop yields
Deleting sensitive data
Lowering product quality
A notable characteristic of Data Science in the modern era is:
Lack of data
Information abundance and the need for smart processing tools
Decreased demand for analysis
No need for statistics
Big Data is often described by the 3Vs. Which of the following is not one of the original 3Vs?
Volume
Velocity
Variety
Value
Which is an example of Data Science in e-commerce?
Personalized product recommendations
Storing website pages
Encoding HTML data
Random advertising
What is the primary role of a Data Scientist?
Manually collect data
Build models and derive insights
Write technical manuals
Draw network diagrams
Which factor distinguishes modern Data Science from traditional approaches?
Relying entirely on intuition
Using advanced algorithms and statistical models
Not needing software
Using paper and pen
What does Data-driven Decision Making mean?
Making decisions based on intuition
Basing decisions on evidence from data
Randomly choosing outcomes
Avoiding analysis
What is a major challenge in Data Science today?
Lack of data
Ensuring privacy and security
Absence of machine learning
Inability to store data
In Data Science, Predictive Analytics refers to:
Analyzing historical data only
Predicting future trends
Describing current data
Classifying data
Which is an example of Data Science in logistics?
Predicting delivery time
Manually sorting packages
Deleting orders
Randomly changing routes
What is the ultimate goal of Data Science?
Collect as much data as possible
Turn data into value and knowledge to support decisions
Store data indefinitely
Reduce processing costs
Structured data is commonly stored in:
.txt files
Relational databases (SQL)
Images and videos
PDF files
Unstructured data typically includes:
CSV files
Excel spreadsheets
Images, videos, and free text
Numeric data
Semi-structured data is:
Data in JSON or XML format
Data in SQL tables
Audio data
Unreadable data
The Data Cleaning process involves:
Collecting data
Processing to remove errors and missing values
Deploying models
Evaluating models
What is an outlier in a dataset?
Data close to the mean
Missing data
Abnormal data that is far from the rest
Duplicate data
Which is an example of discrete data?
Height
Number of students
Temperature
Weight
Which is an example of continuous data?
Eye color
Number of products
Temperature or height
Student ID
Which is a categorical variable?
Age
Satisfaction level: Low, Medium, High
Weight
Revenue
Which is a numerical variable?
Gender
Occupation
Income
Product name
Feature Engineering is the process of:
Collecting new data
Creating new features from raw data
Evaluating models
Deleting data
In Machine Learning, the label is:
An attribute that describes the data
The target variable to be predicted
The test dataset
A random variable
Data normalization aims to:
Increase data size
Standardize data to the same scale
Delete all data
Split data into many tables
One-Hot Encoding is used to:
Convert variables to strings
Encode categorical variables into binary vectors
Remove unnecessary variables
Shuffle the data
Exploratory Data Analysis helps you:
Predict the future
Understand features and relationships among variables
Increase processing speed
Reduce the amount of data
Missing data is a common issue and is often handled by:
Removing all data
Imputing with mean, median, or predictive models
Assigning entirely different values
Deleting the entire dataset
Time-series data is characterized by:
No inherent order
A time factor associated with each record
No classification
Only categorical values
Data Integration is the process of:
Combining data from multiple sources
Splitting data into smaller parts
Data Pipeline is used for:
Manual storage
Automating the collection, processing, and storage of data
Copying data
Visualization
Big Data is characterized by:
Small data volume
Easy processing with Excel
Large volume, velocity, and variety of data
Impossible to store
The ultimate goal of data processing is:
Having as much data as possible
Accurate, usable data for analysis and modeling
Reducing the amount of data
Deleting erroneous data
The first step in the Data Science process is:
Data collection
Defining business goals and problems
Model training
Model deployment
Data Cleaning helps to:
Make data clean by removing missing and duplicate records
Reduce file size
Change the data format
Insert data into a report
Exploratory Data Analysis (EDA) is:
Surveying and visualizing to understand the data
Predicting future values
Storing data
Training models
Feature Selection is the step that:
Removes irrelevant or redundant variables
Adds new data
Predicts labels
Standardizes data
Train/Test Split is used to:
Separate data into training and testing sets
Copy data
Increase dataset size
Randomly shuffle data
Model Training is:
The process of teaching a model to learn from data
Deleting erroneous data
Saving the model to a file
Storing data
Supervised Learning requires:
Labeled data
Unlabeled data
Random data
Noisy data
Unsupervised Learning is used when:
Labeled data is available
Labels are not available and hidden structure is sought
Data is erroneous
Data is small
An example of an Unsupervised Learning algorithm is:
Linear Regression
Decision Tree
K-Means Clustering
Logistic Regression
Model Evaluation aims to:
Assess model performance on unseen data
Reduce the size of data
Increase data
Delete variables
Cross Validation helps to:
Stably evaluate a model by splitting data into multiple folds
Train faster
Increase bias
Reduce the number of features
Overfitting occurs when:
A model learns the training data too well and performs poorly on the test set
The model has not learned enough
The data is faulty
The algorithm is wrong
Underfitting occurs when:
The model is too complex
The model is too simple and fails to learn the patterns
The data is faulty
There is excessive data
Hyperparameter Tuning is:
Optimizing the model’s controlling parameters
Collecting data
Reducing the number of samples
Increasing model size
After a model performs well, the next step is:
Delete the model
Deployment
Stop using it
Cease analysis
Model Monitoring aims to:
Track model performance after deployment
Delete faulty models
Speed up processing
Shut down the entire system
A Feedback Loop in Data Science is:
A cycle of updating the model from real-world feedback
Unlimited data replication
Deleting old data
Stopping the model
Data Visualization is typically performed during:
EDA and result reporting
Data collection
Model training
Deployment
The ultimate result of the Data Science process is:
An effective model that supports real-world decision-making
Simple charts only
Raw data
An Excel file
The most popular programming language in Data Science is:
Java
Python
C++
PHP
Besides Python, a language widely used in statistical analysis is:
R
C#
Kotlin
Swift
SQL is used to:
Train models
Query and manage relational databases
Draw charts
Create PowerPoint reports
The Python library specialized for tabular data processing is:
NumPy
Matplotlib
Pandas
TensorFlow
The Python library for basic Machine Learning is:
Seaborn
Scikit-learn
Plotly
PySpark
TensorFlow is a framework widely used for:
Database administration
Deep Learning
Statistical analysis
Data storage
PyTorch is developed by:
Microsoft
Amazon
Meta (Facebook)
IBM
NumPy is primarily used for:
Charting
Numerical array processing and matrix computations
Building web applications
Data storage
Matplotlib is used for:
Model training
Charting and data visualization
Dashboard creation
Database creation
Seaborn is:
An extension of Matplotlib for more aesthetic statistical plots
A data cleaning tool
A web scraping tool
A data browser
Power BI and Tableau are tools for:
Database administration
Data visualization and reporting
Writing Python code
Data storage
Jupyter Notebook is commonly used to:
Write code and present interactive data analysis
Run office software
Store databases
Host a web server
Google Colab is:
An offline programming tool
Google’s free online notebook environment
A database management system
Visualization software
Apache Spark is a tool for:
Storing static data
Distributed processing of Big Data
Creating dashboards
Compressing data
Hadoop is used for:
Storing and processing big data using the MapReduce model
Image analysis
Designing neural networks
Building web applications
In Data Science, Docker is used to:
Write Python code
Package and deploy model environments
Save Excel files
Create reports
Apache Airflow is used to:
Schedule and automate data workflows
Manage databases
Create charts
Train models
MLflow is used to:
Track experiments and manage machine learning models
Visualize data
Write SQL code
Store JSON files
Which of the following cloud platforms is not a primary tool for Data Science?
Google Cloud
AWS
Microsoft Azure
Canva
In Data Science, the purpose of a dashboard is:
Store data
Visualize and monitor performance metrics
Train models
Clean data
