Chapter 4: Data Science and Statistical Modeling

1. Supervised learning is one of the main category of __

a) Statistical modeling

b) Data wrangling

c) Data collection

d) Evaluation modeling

Answer: a) Statistical modeling

2. For accurate data collection, an existing data from __ is acquired.

a) Sampling department

b) Modeling techniques

c) Meteorological department

d) Educational departments

Answer: a) Sampling department

3. Natural disaster prediction helps the people to take preventive measures against

__

a) Pandemic

b) Cyclones

c) Data analysis

d) Temperature

Answer: b) Cyclones

4. A linear regression model is a mathematical equation that allows us to predict a response

for a given __

a) Predictor value

b) Variable

c) Intercepts

d) Model

Answer: a) Predictor value

5. A point where graph line meet the y-axis is called __

a) Y-axis

b) Y-intercept

c) Y-coordinates

d) Y-dimension

Answer: b) Y-intercept

6. Clustering algorithms are examples of __

a) Classical learning

b) Professional learning

c) Unsupervised learning

d) Cognitive learning

Answer: c) Unsupervised learning

7. We can create our own database by generating some random number by using

__

a) Python code

b) Data code

c) System code

d) Pseudo code

Answer: a) Python code

8. During summer people of hot regions consume more cold items. This information is

called __

a) Parameter

b) Population

c) Statistics

d) Random sample

Answer: c) Statistics

9. If there are some flaws in data collection method, after analyzing such data the result

would be more __

a) Accurate

b) Huge

c) No result can be drawn

d) Erroneous

Answer: d) Erroneous

10. Experimental design is a tool used to __ the result effectively.

a) Add

b) Evaluate

c) Organize

d) Test

Answer: b) Evaluate

1. List out the parameters and statistics from given statements:

a) Average height of a giraffe.

Answer: Statistic (calculated from a sample).

b) Average weight of watermelon.

Answer: Statistic (calculated from a sample).

c) There are 430 doctors in a hospital.

Answer: Parameter (a fixed, complete count of a specific population).

d) Average age of students of 6th class in a school is 12 years.

Answer: Parameter (a fixed value for a specific group).

2. If we want to collect data about exports from Pakistan in last five years, how libraries can

help you to collect data?

Answer: Libraries can help by providing access to:

Government publications and statistical yearbooks.

Economic surveys and reports from the Ministry of Commerce.

Databases like the World Bank or IMF data portals (accessible via library

subscriptions).

Academic journals and newspapers that report on trade statistics.

3. Make a pie chart of vegetable prices in the market. Consider five to ten vegetables.

Answer: (This is a practical activity. A sample pie chart would show the proportion of

the total cost for each vegetable. For example, if potatoes cost Rs. 50, tomatoes Rs.

80, onions Rs. 60, etc., the pie chart would divide a circle into slices proportional to

these values.)

4. Enlist steps to represent the monthly temperatures of a Pakistan city in 2023 from

January till December using a line graph.

Answer:

1. Collect the monthly temperature data for the chosen city in 2023.

2. Draw the x-axis and label it with the months (Jan, Feb, ..., Dec).

3. Draw the y-axis and label it with the temperature scale (e.g., 0°C to 50°C).

4. Plot a point for each month's average temperature.

5. Connect the points with straight lines.

6. Add a title and label the axes clearly.

5. Define either of supervised learning and unsupervised learning. Give reasons for your

preference for the other.

Answer:

Supervised Learning: The model is trained on labeled data (input-output

pairs) to predict outcomes for new data.

Unsupervised Learning: The model is trained on unlabeled data to find

hidden patterns or groupings.

Preference: I prefer Supervised Learning because it is more straightforward

to evaluate (we know the correct answers) and is highly effective for tasks

like classification and regression where clear labels exist.

6. Write a Python code to generate a dataset with two variables where y = x² + 2x. Fit

scatter plot and box plot on this data.

Answer:

python

import numpy as np

import matplotlib.pyplot as plt

import seaborn as sns

# Generate data

x = np.linspace(-10, 10, 50)

y = x**2 + 2*x

# Scatter plot

plt.figure(figsize=(10, 4))

plt.subplot(1, 2, 1)

plt.scatter(x, y, color='blue')

plt.title('Scatter Plot')

plt.xlabel('x')

plt.ylabel('y = x² + 2x')

# Box plot

plt.subplot(1, 2, 2)

sns.boxplot(y=y, color='green')

plt.title('Box Plot of y')

plt.tight_layout()

plt.show()

7. Relate some real world examples (other than Airbnb, Facebook and YouTube) where data

science was used to improve marketing strategies and enhance the business.

Answer:

Amazon: Uses data science for product recommendations, dynamic pricing,

and supply chain optimization.

Netflix: Uses viewing data to recommend shows and decide which new

content to produce.

Starbucks: Uses customer data and location analytics to decide where to

open new stores and personalize offers.

Spotify: Uses listening data to create personalized playlists (Discover

Weekly) and recommend new artists.

1. Simulate on paper, an experimental design for awareness of food security (Narrative

Visualization).

Answer: (This is a descriptive design task. A simulated design would include:)

Objective: To measure the level of awareness about food security among a

target population.

Target Population: Residents of a specific city or region.

Sampling Method: Stratified random sampling (dividing the population into

groups like income levels, age groups, and randomly selecting participants

from each).

Data Collection Tool: A structured questionnaire with questions on:

Knowledge of food security concepts.

Access to nutritious food.

Perceived barriers to food security.

Data Analysis Plan: Use descriptive statistics (percentages, averages) and

inferential statistics (chi-square tests) to analyze the data.

Visualization: Create bar charts showing awareness levels by demographic

group, pie charts showing sources of information, and a narrative

storyboard to communicate the findings to policymakers.

2. Sketch primary data collection methods in context of disease outbreak, like seasonal flu.

Answer: Primary data collection for a disease outbreak like seasonal flu involves:

Surveys/Questionnaires: Distributed to the public to gather information on

symptoms, vaccination status, and travel history.

Interviews: Conducted with patients and healthcare workers to get detailed

case histories.

Clinical Examinations: Collecting biological samples (swabs, blood) from

suspected cases for laboratory confirmation.

Observational Studies: Monitoring hygiene practices in public places and

healthcare settings.

Contact Tracing: Identifying and interviewing people who have been in

contact with confirmed cases.

Reporting Systems: Establishing a system for healthcare providers to report

new cases to public health authorities.

3. Argue about the use of statistical modeling techniques. Highlight all techniques discussed

in the unit.

Answer: Statistical modeling is crucial for making sense of data and making

predictions. It allows us to:

Identify Relationships: Understand how variables are related (e.g., does

temperature affect ice cream sales?).

Make Predictions: Forecast future trends (e.g., predicting disease spread).

Test Hypotheses: Validate theories with data.

Techniques discussed:

Linear Regression: Models the relationship between a dependent variable

and one or more independent variables.

Classification: Predicts categorical outcomes (e.g., spam vs. not spam).

Clustering: Groups similar data points together (unsupervised learning).

Time Series Analysis: Analyzes data points collected over time to identify

trends and patterns.

4. Compare Linear regression and classification. Emphasize on their respective roles in

statistical modeling.

Answer:

Role in Used for Used for

Modeling forecasting decision-

and making,

understandi pattern

ng recognition

relationshi , and

ps between sorting

variables. data into

predefined

classes.

5. Define either of supervised learning and unsupervised learning. Give reasons for your

preference for the other.

(Note: This is a repeat of SRQ #5. Please refer to the answer provided above.)

6. Write a Python code to generate a dataset with two variables where y = x² + 2x. Fit

scatter plot and box plot on this data.

(Note: This is a repeat of SRQ #6. Please refer to the answer provided above.)

7. Relate some real world examples (other than Airbnb, Facebook and YouTube) where data

science was used to improve marketing strategies and enhance the business.

(Note: This is a repeat of SRQ #7. Please refer to the answer provided above.)

(These activities are from the book text, separated as requested.)

Activity 1

Practice all the Python code given within this Unit.

Objective: To help students develop the skill of data analysis by using Python.

Outcome: This activity will help students understand the basic syntax of Python

for data analysis.

Activity 2

Data collection by Interview method:

Objective: To help students understand the process of data collection through

survey and questionnaires.

Instructions:

Divide the students into small groups (not more than 3 students in each

group).

Ask the students to collect data by interviewing their fellow students,

about their academic performance, study schedule, outdoor games etc.

Ask them to organize data and plot them by using their favorite tool.

Supervise them to draw conclusions about correlation among various

factors and academic performance.

Various groups will compare their results/conclusions and find the

similarities and differences.

Outcome: This activity will help students understand positive or negative

correlation among academics and not academic activities. They will also develop

the skill of conduction interview as an interviewer as well as interviewee.

Activity 3

Build more statistical models on new datasets.

Objective: To help students develop the skill of data analysis by using Python and

other tools.

Instructions:

Ask the students to search and download some dataset of their choice

from the internet.

Perform the steps of data cleaning and feature selection.

Visualize the data and create various graphs according to the nature of

dataset. Students can use other tools like MS-Excel, Weka, Pytorch, R

studio etc.

Supervise and help them finding easy and simple tutorials on internet,

students can later share small video tutorial through Whatsapp.

Outcome: This activity will help students learn new tools at their own.Chapter

Guide: Emerging Technologies

1. Fluid flow sensor is a type of:

a) Environmental sensor

b) Industrial sensor

c) Motion detection sensor

d) Climate control sensor

Answer: b) Industrial sensor

2. The cloud model that provides services to multiple organizations is known as:

a) Public cloud

b) Private cloud

c) Community cloud

d) Hybrid cloud

Answer: a) Public cloud

3. The cloud model that provides best security of data is:

a) Public cloud

b) Private cloud

c) Community cloud

d) Hybrid cloud

Answer: b) Private cloud

4. Exchange of information between IoT devices and end-users is provided by:

a) Cloud computing

b) Big data analytics

c) Embedded systems

d) Communication protocols

Answer: d) Communication protocols

5. The Blockchain network that is most difficult to set up is:

a) Public Blockchain network

b) Private Blockchain network

c) Permissioned Blockchain network

d) Consortium Blockchain network

Answer: a) Public Blockchain network

6. The Blockchain network that is open for everyone to join is called:

a) Public Blockchain network

b) Private Blockchain network

c) Permissioned Blockchain network

d) Consortium Blockchain network

Answer: a) Public Blockchain network

7. The sensors used to prevent unauthorized movement in restricted areas are called:

a) Environment sensors

b) Motion detection sensors

c) Industrial sensors

d) Restricted area sensors

Answer: b) Motion detection sensors

8. Sensors used for measuring humidity, air quality, air pressure, temperature and wind

speed are called:

a) Environment sensors

b) Industrial sensors

c) Motion detection sensors

d) Measurement sensors

Answer: a) Environment sensors

9. Software that is permanently installed in an electronic machine is known as:

a) System software

b) Application software

c) Machine software

d) Firmware

Answer: d) Firmware

10. A digital ledger of transactions shared and maintained by network users is called:

a) Database

b) Accounting software

c) Blockchain

d) Network ledger

Answer: c) Blockchain

1. How IoT can enhance our daily life?

Answer: IoT enhances daily life by automating and optimizing routine tasks.

Examples include smart home devices that control lighting and temperature, wearable

fitness trackers that monitor health, smart appliances that reorder groceries, and

connected cars that provide navigation and safety features.

2. Provide 3 examples of WSNs used in IoT system.

Answer:

1. Environmental Monitoring: A network of sensors deployed in a forest to

detect temperature, humidity, and smoke for early fire detection.

2. Smart Agriculture: Sensors in a field to monitor soil moisture, nutrient

levels, and weather conditions to optimize irrigation and fertilization.

3. Industrial Automation: Sensors on factory machinery to monitor vibration,

temperature, and performance for predictive maintenance.

3. Differentiate between public cloud model and private cloud model.

Answer:

Public Cloud: The cloud infrastructure is owned and operated by a third-

party cloud service provider (e.g., AWS, Azure) and is shared by multiple

organizations over the internet. It is cost-effective and scalable but offers

less control and security.

Private Cloud: The cloud infrastructure is provisioned for exclusive use by

a single organization. It can be physically located on the company's on-site

datacenter or hosted by a third-party. It offers greater control, security, and

customization but is more expensive.

4. What is Blockchain? Why data stored in a Blockchain is secure?

Answer: A blockchain is a decentralized, distributed digital ledger that records

transactions across many computers. Data stored in a blockchain is secure because:

Decentralization: There is no single point of failure or control.

Immutability: Once a block is added to the chain, it cannot be altered or

deleted.

Cryptographic Hashing: Each block contains a cryptographic hash of the

previous block, creating a secure chain. Any attempt to alter a block would

change its hash, breaking the chain and alerting the network.

Consensus Mechanisms: Transactions are validated by network participants

before being added.

5. Why integration of Blockchain and IoT is beneficial?

Answer: Integrating Blockchain with IoT provides:

Enhanced Security: Blockchain's immutable ledger protects IoT data from

tampering.

Decentralized Trust: Eliminates the need for a central authority to manage

IoT devices.

Automated Transactions: Smart contracts can automate actions between IoT

devices (e.g., a device paying for data).

Data Integrity: Ensures that data collected by IoT sensors is accurate and

unaltered.

6. Define permissioned Blockchain network.

Answer: A permissioned blockchain network is a type of blockchain where access to

the network is restricted. Only authorized participants who have been granted

permission can join the network, view the ledger, and participate in the consensus

process.

1. What is Blockchain technology? Describe in detail how transactions are processed using

Blockchain technology.

Answer: Blockchain technology is a decentralized, distributed, and often public

digital ledger used to record transactions across many computers. It is the foundation

of cryptocurrencies like Bitcoin.

How Transactions are Processed:

1. Initiation: A user initiates a transaction (e.g., sending cryptocurrency).

2. Broadcasting: The transaction is broadcast to a peer-to-peer network of

nodes.

3. Validation: Network nodes validate the transaction using consensus

mechanisms (like Proof of Work or Proof of Stake) to ensure the sender has

the funds and the transaction is legitimate.

4. Block Creation: Validated transactions are grouped into a "block."

5. Hashing: The block is given a unique cryptographic hash, which includes

the hash of the previous block, linking them together.

6. Addition to Chain: The new block is added to the blockchain.

7. Confirmation: The transaction is now permanently recorded and visible to

all participants.

2. Enlist the three key technologies that enabled IoT.

Answer: The three key technologies that enabled IoT are:

1. Communication Protocols: Technologies like Wi-Fi, Bluetooth, Zigbee, and

5G allow devices to connect and exchange data.

2. Cloud Computing: Provides the storage and processing power needed to

handle the massive amounts of data generated by IoT devices.

3. Embedded Systems: Small, low-power computers (like microcontrollers)

that are integrated into everyday objects, enabling them to sense, process,

and communicate.

3. Criticize the negative impacts of AI systems in the domain of education and learning of

students.

Answer: While AI has benefits, it also has negative impacts on education:

Over-reliance: Students may become overly dependent on AI tools (like

ChatGPT) for answers, hindering critical thinking and problem-solving

skills.

Bias: AI algorithms can perpetuate biases present in their training data,

leading to unfair grading or biased learning materials.

Privacy Concerns: AI systems collect vast amounts of student data, raising

concerns about data privacy and security.

Dehumanization: Overuse of AI can reduce human interaction between

teachers and students, which is crucial for mentorship and social-emotional

learning.

Digital Divide: Students without access to AI tools may be left behind,

widening the educational gap.

4. Examine the reasons behind the conflicting requirements among stakeholders during the

development of AI systems.

Answer: Conflicting requirements among stakeholders arise due to differing

priorities:

Users: Want AI that is easy to use, accurate, and respects their privacy.

Developers: Want to build powerful, innovative systems that are technically

feasible and profitable.

Regulators: Want to ensure AI is safe, ethical, and does not harm society.

Business Owners: Want AI that maximizes profits, reduces costs, and

provides a competitive advantage.

Ethicists: Want AI that is fair, transparent, and does not perpetuate bias.

These conflicting goals (e.g., profit vs. privacy, innovation vs. safety) create

challenges in the design and deployment of AI systems.

5. Consider creating a cutting-edge system for language learning. The priorities of teachers,

learners, and programmers will all differ. How would it help to make the new language

learning system better by incorporating the varying priorities? How can AI be added to it?

Answer:

Teachers' Priorities: They want a system that supports pedagogy, tracks

student progress, and provides insights into learning gaps.

Learners' Priorities: They want a system that is engaging, personalized, and

allows them to practice at their own pace.

Programmers' Priorities: They want a system that is scalable, maintainable,

and uses efficient algorithms.

How incorporating priorities helps:

A system that includes teacher dashboards, student gamification, and robust

backend architecture will be more effective and widely adopted.

It ensures the system is pedagogically sound, user-friendly, and technically

reliable.

Adding AI:

Personalized Learning Paths: AI can analyze student performance and adapt

the curriculum to their needs.

Intelligent Tutoring: AI chatbots can provide instant feedback and practice

conversations.

Speech Recognition: AI can assess pronunciation and provide corrections.

Content Generation: AI can generate customized reading materials and

exercises.

(These activities are from the book text, separated as requested.)

Activity 1

Take an example of an IoT application and identify how each component of connectivity,

processing, AI, cloud computing, and battery power enables that application.

Answer: (Example: Smart Home Security System)

Connectivity: Wi-Fi or Zigbee connects cameras, door sensors, and smart

locks to the home network.

Processing: A local hub or the camera itself processes video feeds to detect

motion.

AI: AI algorithms distinguish between a person, a pet, or a car, reducing

false alarms.

Cloud Computing: Video footage is stored in the cloud for remote access

and backup.

Battery Power: Battery-powered sensors allow for flexible placement

without wiring.

Activity 2

Have the students work through a Blockchain application where the students act like

Blockchain users and maintain a ledger through their peer-to-peer network in class.

Answer: (Classroom Simulation Guide)

Setup: Assign each student a "wallet" with a starting balance. Designate

some students as "miners."

Transaction: One student announces a transaction (e.g., "I am sending 5

coins to Student B").

Broadcast: The transaction is written on the board for all to see.

Validation: Miners verify the sender has enough coins.

Block Creation: Miners group transactions into a "block."

Hashing: The block is given a unique number (hash) and linked to the

previous block.

Ledger Update: All students update their personal ledgers with the new

block.

Outcome: Demonstrates decentralization, immutability, and consensus.

Activity 3

Have a discussion in class where you divide the class into groups. Each group is further

divided into three roles where one role represents advertisers that are trying to sell their

products on a social media platform, the second role is using social media to connect with

friends, and the third role is the developers who is trying to sign up more users and trying to

get the existing users to spend more time on their application. Help the discussion on how

each role has different expectations from the AI algorithms being used in the application.

Answer: (Discussion Guide)

Advertisers: Expect AI to target ads precisely to users most likely to buy,

maximizing return on investment.

Users: Expect AI to show relevant content and connections, while

respecting their privacy and not being manipulative.

Developers: Expect AI to maximize user engagement, time spent, and user

growth, often prioritizing these metrics over user well-being.

Discussion Points: How do these expectations conflict? (e.g., Advertisers

want more ads, users want fewer). How can AI be designed to balance these

competing interests ethically?