PART 1

👉 Question Number: 1
👉 Unit Number: 1
Question: Explain in detail about data analytics life cycle

👉 Answer:
Definition:

As referenced in Geeks for Geeks, the Data Analytics Life Cycle is a structured framework that
outlines the roadmap of how data is generated, collected, processed, analyzed, and visualized
to solve a specific business problem and drive strategic decision-making.

Explanation:

The Data Analytics Life Cycle generally consists of six major phases:

●​ Phase 1: Discovery: The team learns the business domain, assesses the resources
available (people, technology, time), and formulates the primary business problem into a
strategic analytics goal.
●​ Phase 2: Data Preparation: This is the most time-consuming phase. It involves extracting
data from various sources, cleaning it (handling missing values and anomalies), and
transforming it into a structured format suitable for analysis.
●​ Phase 3: Model Planning: The team determines the methods, techniques, and workflow
to build the model. They select algorithms (like regression or clustering) based on the
data variables and business goals.
●​ Phase 4: Model Building: The analytical models are developed and executed on the
prepared datasets. The data is usually split into training and testing sets to evaluate the
model's accuracy.
●​ Phase 5: Communicate Results: The findings are interpreted and presented to the
business stakeholders. Data visualization tools are used to showcase whether the project
succeeded or failed in achieving its initial objectives.
●​ Phase 6: Operationalize: The final model is deployed into a real-time production
environment to deliver ongoing business value.

Real-World Example (Amazon):

In the context of the Amazon Product Recommendation System:

●​ Discovery: Amazon sets a goal to increase cross-selling revenue.(users who buy

smartphones also buy phone cases)
●​ Data Preparation: Data engineers gather user browsing history, purchase logs, and cart
data, cleaning out incomplete entries.
●​ Model Planning: Data scientists decide to use collaborative filtering algorithms.
 ●​ Model Building: The model is trained on past user behavior to find patterns (e.g., users
who buy smartphones also buy phone cases).
●​ Communicate Results: The team presents a dashboard showing a potential 15% increase
in sales to Amazon executives.
●​ Operationalize: The recommendation engine is integrated into the live Amazon website,
showing users the "Customers who bought this item also bought" section.

👉 Question Number: 2
👉 Unit Number: 1
Question: Explain in short about application of data analytics across different industries

👉 Answer:
Definition:

As highlighted by Geeks for Geeks application case studies, Data Analytics applications refer to
the practical implementation of statistical algorithms, machine learning, and data modeling
techniques to solve domain-specific challenges, optimize business operations, and create new
revenue streams across various sectors.

Explanation:

Data Analytics is transforming multiple industries by providing actionable insights:

●​ Retail and E-commerce: Used for inventory forecasting, dynamic pricing, and
personalized customer recommendations to enhance the shopping experience and
boost sales.
●​ Healthcare: Applied in predicting disease outbreaks, optimizing hospital staffing,
improving patient care through personalized medicine, and accelerating drug discovery.
●​ Banking and Finance: Essential for detecting fraudulent transactions in real-time,
assessing credit risk for loan approvals, and algorithmic trading.
●​ Logistics and Supply Chain: Utilized for route optimization, fleet management, and
predicting shipping delays to ensure timely deliveries and reduce fuel consumption.
●​ Entertainment and Media: Used to analyze viewer preferences, optimize content
streaming quality, and recommend movies or music to keep users engaged.

Real-World Example (Amazon):

●​ E-commerce: Amazon uses analytics to dynamically change product prices millions of

times a day based on competitor pricing and demand.
●​ Logistics: Amazon Prime utilizes route optimization analytics to ensure drivers take the
fastest, most fuel-efficient paths to achieve 1-day or same-day delivery.
●​ Entertainment: Amazon Prime Video analyzes which scenes users rewind or skip to
 recommend the exact genres and movies they want to watch next.

👉 Question Number: 3
👉 Unit Number: 1
Question: Explain in detail about data analytics with examples

👉 Answer:
Definition:

According to Geeks for Geeks, Data Analytics is the scientific process of collecting, inspecting,
cleansing, transforming, and modeling raw datasets to discover hidden patterns, unseen
trends, and correlations. The ultimate goal is to draw meaningful conclusions that support
strategic business decision-making.

Explanation:

Data Analytics is broadly categorized into four main types, each answering a specific business
question:

●​ Descriptive Analytics (What happened?): Summarizes historical data to understand

current or past business performance. It uses basic math like averages and percentiles.
●​ Diagnostic Analytics (Why did it happen?): Dives deeper into descriptive data to find
the root causes of specific trends or anomalies. It involves data discovery and drill-down
techniques.
●​ Predictive Analytics (What is likely to happen?): Uses historical data, statistical
algorithms, and machine learning to forecast future outcomes and trends.
●​ Prescriptive Analytics (What should we do?): The most advanced stage, which
recommends specific actions to take based on the predicted outcomes to maximize
business value.

Real-World Example (Amazon):

Using the Amazon Prime ecosystem:

●​ Descriptive Analytics: Amazon generates a monthly report showing that sales of winter
jackets increased by 40% in November.
●​ Diagnostic Analytics: Analysts drill down into the data and discover that the spike in
sales was due to a sudden drop in temperature combined with a successful targeted
email campaign.
●​ Predictive Analytics: Based on past years' data and upcoming weather forecasts,
Amazon predicts that they will need 200,000 winter jackets in their New York fulfillment
centers by December 1st.
●​ Prescriptive Analytics: The analytics system automatically generates purchase orders to
 suppliers and optimally routes the inventory to specific regional warehouses to prevent
stockouts and minimize shipping times.
PART 2

👉 Question Number: 4
👉 Unit Number: 1
Question: Differentiate data analytics from data analysis - with minimum 10 differences

👉 Answer:
Definition:

As per Geeks for Geeks, both terms are often used interchangeably but hold distinct
meanings:

●​ Data Analysis Workflow: Examines Past Data ➔ Cleans & Organizes ➔ Identifies Trends
➔ Answers "What happened?"
●​ Data Analytics Workflow: Encompasses Analysis ➔ Applies Machine Learning ➔
Predicts Future Trends ➔ Answers "What should we do next?"

Explanation (10 Differences):

Feature Data Analysis Data Analytics

1. Scope Micro-level (focused on Macro-level (a broad,

specific, historical datasets). overarching science that
includes analysis).

2. Time Focus Looks backward at historical Looks forward to forecast

data (Past-oriented). trends (Future-oriented).

3. Core Goal To uncover patterns and To optimize systems, make

understand current business predictions, and drive
 performance. automated decisions.

4. Key Questions "What happened?" and "Why "What will happen next?" and
did it happen?" "How can we influence it?"

5. Process Flow Querying ➔ Cleaning ➔ Data Modeling ➔ Algorithm

Aggregating ➔ Visualizing Training ➔ Predicting ➔
Automating

6. Output Dashboards, static reports, Predictive models,

and business summaries. recommendation engines, and
automated workflows.

7. Tools Used Excel, SQL, Tableau, PowerBI. Python, R, TensorFlow, Hadoop,

Advanced Machine Learning
libraries.

8. Types Descriptive and Diagnostic Predictive and Prescriptive

Analysis. Analytics.

9. Complexity Moderate (requires statistical High (requires programming,

and querying skills). statistics, and machine learning
skills).
10. Practitioner Data Analyst, Business Data Scientist, Data Engineer,
Analyst. Analytics Professional.

Real-World Example (Amazon):

●​ Data Analysis: An Amazon Business Analyst uses SQL to find out that sales of "Kindle
Paperwhite" dropped by 15% last August.
●​ Data Analytics: An Amazon Data Scientist builds a machine learning model that predicts
next August's Kindle sales based on weather, marketing spend, and competitor pricing,
automatically adjusting the price to maximize profit.

👉 Question Number: 5
👉 Unit Number: 1
Question: Explain in detail about significance of data analytics in IOT Domain

👉 Answer:
Definition:

The Internet of Things (IoT) refers to physical devices connected to the internet, collecting and
sharing data. As referenced in Geeks for Geeks, Data Analytics acts as the "brain" of IoT.

●​ The IoT Analytics Flow: IoT Sensors Collect Raw Data ➔ Cloud Storage ➔ Data
Analytics Processing ➔ Actionable Insights ➔ Automated Machine Response

Explanation (Significance in Points):

●​ Volume & Velocity Handling: IoT devices generate terabytes of data per second.
Analytics platforms are significant because they can ingest, filter, and process this
massive streaming data in real-time.
●​ Predictive Maintenance: Analytics constantly evaluates sensor data (like temperature or
vibration) to predict when a machine is likely to break down, preventing costly factory
halts.
●​ Cost Optimization: By analyzing energy consumption patterns from smart meters,
businesses can automate heating and lighting, significantly reducing operational costs.
●​ Enhanced Customer Experience: Wearable IoT devices track user health metrics, and
analytics personalize fitness goals or alert doctors in case of heart anomalies.
●​ Edge Analytics: Instead of sending all data to the cloud, analytics can now happen
directly on the IoT device (the "edge"), allowing for split-second autonomous decisions
 without internet latency.

Real-World Example (Amazon):

In Amazon Fulfillment Centers, thousands of Kiva robots (IoT devices) move inventory.

●​ Sensors track robot battery & wheel friction ➔ Analytics processes this data instantly ➔
Predicts a wheel failure ➔ Automatically routes the robot to the maintenance bay before
it breaks down and blocks a busy aisle.

👉 Question Number: 6
👉 Unit Number: 2
Question: Mention the various data sources and collection methods with examples

👉 Answer:
Definition:

According to Geeks for Geeks, a Data Source is the original location where data is born or
stored, while a Collection Method is the technical or manual process used to extract that data.

●​ The Collection Flow: Identify Source ➔ Choose Method (e.g., API/Scraping) ➔ Extract
➔ Store in Database

Explanation (Sources & Methods):

1. Data Sources:

●​ Internal Sources: Data generated within the organization. (e.g., CRM systems, HR
databases, Sales transactional databases).
●​ External Sources: Data gathered from outside the organization. (e.g., Government
census portals, public financial datasets, social media platforms).

2. Data Collection Methods:

●​ Web Scraping: Using software bots to extract large amounts of unstructured data
directly from website HTML.
●​ APIs (Application Programming Interfaces): A secure, structured method where
computers talk to each other to request and deliver specific data in JSON format.
●​ Transactional Tracking: Automatically recording user behavior (clicks, purchases, time
spent) using web cookies and backend server logs.
●​ Surveys & Questionnaires: Manual collection of qualitative and quantitative data directly
from users.

Real-World Example (Amazon):

 ●​ Data Source: User's web browser and Amazon's internal inventory database.
●​ Collection Method: Transactional tracking via cookies collects every product you click
on. Amazon also uses APIs to fetch real-time shipping rates from FedEx or UPS to display
at your checkout screen.

Python Code (Data Collection via Pandas):

As this is a Unit 2 data handling concept, here is a basic implementation of collecting data from
different sources.

Python

# Step 1: Import necessary library for data manipulation​

import pandas as pd​
​
# Step 2: Collecting data from a local CSV file (Internal Source)​
# This simulates reading Amazon's internal sales records​
csv_data = pd.read_csv('amazon_sales_data.csv')​
print("--- Data Collected from CSV ---")​
print(csv_data.head(2))​
​
# Step 3: Collecting data from a Web URL/API (External Source)​
# Pandas can directly scrape tables from HTML pages​
url = "[Link]
html_data = pd.read_html(url)​
print("\n--- Data Collected from Web (HTML) ---")​
print(html_data[0].head(2))​
​
# Step 4: Collecting data from a JSON file (Structured Web Data)​
# This simulates collecting product reviews via an API response​
json_data = pd.read_json('amazon_reviews.json')​
print("\n--- Data Collected from JSON ---")​
print(json_data.head(2))​

Code Explanation:

●​ import pandas as pd: Brings in the primary Python library used for data analytics and
collection.
●​ pd.read_csv(): Method used to extract structured tabular data from a comma-separated
values file (common for internal databases).
 ●​ pd.read_html(): Method that acts as a basic web scraper, automatically finding and
extracting HTML tables from a given website link.
●​ pd.read_json(): Method used to parse JSON formatted data, which is the standard format
for collecting data via modern web APIs.
PART 3

👉 Question Number: 7
👉 Unit Number: 2
Question: Explain the API (APPLICATION PROGRAMMING INTERFACE) and how it is going
to collect the data with example

👉 Answer:
Definition:

As referenced in Geeks for Geeks, an API (Application Programming Interface) is a software

intermediary that allows two different applications to communicate with each other. It acts as a
messenger that takes a request from a user, tells a system what to do, and returns the
response back to the user.

Explanation (Data Collection Flow):

Instead of writing complex web scrapers, analysts use APIs to collect data directly in a clean,
structured format (usually JSON). The flow can be remembered as:

Client Sends Request ➔ API Receives Request ➔ Database Searched ➔ Data Packaged
(JSON) ➔ API Delivers Response to Client

●​ Endpoint: The specific URL where the API can be accessed.

●​ Request (GET): The command used to ask the API to retrieve specific data.
●​ Authentication (API Key): A secure password to ensure only authorized users can
collect the data.
●​ Response: The data returned, which is highly structured and ready for immediate
analysis.

Real-World Example (Amazon):

When you use a third-party price-tracking app (like CamelCamelCamel) to check Amazon
prices:

●​ The app does not guess the price. Instead, it sends a request to the Amazon Product
API.
●​ The API fetches the exact current price of a "Kindle Paperwhite" from Amazon's database.
●​ The API returns this data instantly to the tracking app to display to the user.
Python Code (Data Collection via API):

Python

# Step 1: Import necessary libraries​

import requests​
import pandas as pd​
​
# Step 2: Define the API endpoint URL (Mock Amazon API for example)​
api_url = "[Link]
​
# Step 3: Send a GET request to collect data​
response = [Link](api_url)​
​
# Step 4: Convert the JSON response into a Python dictionary​
data = [Link]()​
​
# Step 5: Convert to a Pandas DataFrame for analytics​
df = [Link]([data])​
print(df[['Product_Name', 'Price', 'Stock_Status']])​

Code Explanation:

●​ import requests: Imports the standard Python library used for making HTTP requests to
APIs.
●​ [Link](api_url): Instructs the program to hit the API link and ask for the data.
●​ [Link](): Automatically parses the structured text data returned by the API into a
usable Python dictionary format.
●​ [Link]([data]): Converts the raw JSON data into a neat, tabular format (rows and
columns) for easy data analysis.

👉 Question Number: 8
👉 Unit Number: 2
Question: What do you mean by data cleaning and how it is processed with python
programming

👉 Answer:
Definition:
Data Cleaning (or data cleansing) is the process of identifying and correcting (or removing)
errors, inconsistencies, missing values, and duplicates in a dataset to improve its quality for
analytics.

Explanation (Data Cleaning Process Flow):

To easily remember the steps, think of this flow:

Raw Data ➔ Remove Duplicates ➔ Handle Missing Values ➔ Fix Formatting/Typos ➔ Remove
Outliers ➔ Clean Data Ready for Modeling

●​ Handling Missing Data: Filling in blank spaces with averages (mean/median) or dropping
the incomplete rows entirely.
●​ Removing Duplicates: Deleting identical repeated entries that could skew the analysis.
●​ Standardizing Formatting: Ensuring all text is consistent (e.g., converting all dates to
YYYY-MM-DD format or all text to lowercase).
●​ Handling Outliers: Identifying extreme values that don't make sense (e.g., a customer
age of 999 years) and correcting or removing them.

Real-World Example (Amazon):

Amazon collects millions of product reviews daily.

●​ The Problem: Some users accidentally submit the same review twice (duplicates), or
leave a star rating but forget to write text (missing values).
●​ The Fix: Amazon's data analysts write Python scripts to automatically drop the duplicate
reviews and fill empty text fields with "No comment provided" before feeding the data
into their sentiment analysis models.

Python Code (Data Cleaning):

Python

# Step 1: Import pandas and numpy​

import pandas as pd​
import numpy as np​
​
# Step 2: Create a sample messy dataset (Amazon Reviews)​
raw_data = {​
'UserID': [101, 102, 102, 104, 105], # Notice 102 is duplicated​
'Rating': [5, 4, 4, [Link], 2], # Notice NaN (missing value)​
'Review': ['Great!', 'Good', 'Good', 'Okay', 'Bad']​
}​
df = [Link](raw_data)​
print("--- Raw Data --- \n", df)​
​
# Step 3: Remove duplicate rows​
df_cleaned = df.drop_duplicates()​
​
# Step 4: Handle missing values (Fill NaN with average rating of 3)​
df_cleaned['Rating'] = df_cleaned['Rating'].fillna(3)​
​
print("\n--- Cleaned Data --- \n", df_cleaned)​

Code Explanation:

●​ [Link]: Represents a missing or blank value in the dataset.

●​ drop_duplicates(): A pandas function that scans the dataset and deletes any perfectly
identical rows.
●​ fillna(3): A pandas function that finds any missing values (NaN) in the 'Rating' column and
replaces them with the number 3, ensuring the math models don't crash later.

👉 Question Number: 9
👉 Unit Number: 2
Question: What do you mean by data quality and what are the significance of it

👉 Answer:
Definition:

As per Geeks for Geeks, Data Quality is a measure of the condition of a dataset based on
factors like accuracy, completeness, consistency, reliability, and timeliness. It dictates whether
the data is fit for its intended purpose.

Explanation (Significance of Data Quality):

High-quality data is the foundation of accurate analytics. Poor data leads to a concept known
as GIGO (Garbage In ➔ Garbage Out).

Key Pillars of Data Quality (Flowchart Style):

Accurate (No errors) + Complete (No blanks) + Consistent (Same format) + Timely (Up-to-date)
= High Data Quality

●​ Significance 1: Reliable Decision Making: High-quality data ensures that business

executives are basing their financial and strategic decisions on facts rather than flawed
 information.
●​ Significance 2: Improved Customer Experience: Accurate data ensures customers are
targeted with relevant ads and receive their products at the correct addresses.
●​ Significance 3: Cost Reduction: Fixing errors early prevents the massive costs
associated with failed marketing campaigns or shipping logistics errors caused by bad
data.
●​ Significance 4: Regulatory Compliance: Maintaining accurate and secure data helps
companies avoid legal penalties regarding data privacy laws.

Real-World Example (Amazon):

●​ Poor Data Quality: If an Amazon customer's address data is inconsistent or missing a zip
code, the package will be delayed, leading to shipping refunds, lost money, and an angry
customer.
●​ High Data Quality: If Amazon's recommendation engine has accurate, complete data on
a user's past purchases, it will successfully recommend a highly relevant product,
resulting in an immediate sale and increased profit.

Python Code (Checking Data Quality):

Python

import pandas as pd​

​
# Step 1: Sample dataset representing Amazon customer profiles​
data = {'CustomerID': [1, 2, 3], 'Age': [25, -5, 30], 'Email': ['a@[Link]', None, 'c@[Link]']}​
df = [Link](data)​
​
# Step 2: Quality Check - Check for missing values (Completeness)​
missing_data = [Link]().sum()​
print("Missing Values Check:\n", missing_data)​
​
# Step 3: Quality Check - Check for logical errors (Accuracy)​
# Age cannot be negative. This flags bad quality data.​
bad_age_check = df[df['Age'] < 0]​
print("\nInvalid Age Check (Inaccurate Data):\n", bad_age_check)​

Code Explanation:

●​ [Link]().sum(): This is a standard data quality check in pandas that scans every column
 and counts exactly how many missing/blank values exist.
●​ df[df['Age'] < 0]: This is a logical rule check to ensure data accuracy by identifying rows
where the age is an impossible negative number.
👉 Unit Number: 2
Question: Differentiate database from the data warehouse

👉 Answer:
Definition:

According to Geeks for Geeks, a Database is a collection of current, operational data designed
to process quick, day-to-day transactions (OLTP). A Data Warehouse is a massive, centralized
repository that stores historical, aggregated data from multiple sources specifically designed for
complex data analytics and reporting (OLAP).

Explanation (Data Flow):

App/User Action ➔ Database (Quick Save) ➔ ETL Process (Extract, Transform, Load) ➔ Data
Warehouse (Deep Analytics)

Key Differences:

Feature Database (OLTP) Data Warehouse (OLAP)

1. Primary Transaction processing (running Data analysis (analyzing the

Purpose the business). business).

2. Data Age Current, up-to-date, real-time data. Historical data (months or years
old).

3. Processing Optimized for fast writes (inserting Optimized for fast reads (querying
Speed data). large data).

4. Architecture Normalized (avoids data Denormalized (optimized for fast

Types redundancy). querying).
 5. Primary Users End-users, apps, Database Data Analysts, Data Scientists,
Administrators. Executives.

Real-World Example (Amazon):

●​ Database: When you add a pair of shoes to your Amazon cart and checkout, that
transaction is instantly processed and saved in an Amazon Database to ensure your
payment goes through quickly.
●​ Data Warehouse: Amazon copies that transaction into a massive Data Warehouse
containing 10 years of shoe sales. A data analyst queries this warehouse to find out
which shoe brand sold the most during winters from 2015 to 2025.

👉 Question Number: 11
👉 Unit Number: 2
Question: Explain in detail about structure data with example

👉 Answer:
Definition:

As per Geeks for Geeks, Structured Data is highly organized information that is formatted into
a strictly defined schema, typically consisting of rows and columns. This makes it easily
searchable by algorithms and relational databases.

Explanation (Structure Flow):

Raw Data Input ➔ Define Data Types (Int, String, Float) ➔ Map to Rows
& Columns ➔ Store in SQL Database ➔ Query instantly

Key Characteristics:

●​ Defined Schema: It strictly follows a predefined model. If a column is designed to hold

numbers (like a price), you cannot enter letters into it.
●​ Easy Storage: Because it is so organized, it consumes less storage space compared to
unstructured data like images or videos.
●​ Highly Searchable: It can be instantly queried, filtered, and aggregated using SQL
(Structured Query Language).
●​ Machine Learning Ready: Standard statistical algorithms and machine learning models
are easily applied to this data format.

Real-World Example (Amazon):

An Amazon Customer Profile Table is pure structured data:

 ●​ Row 1: [UserID: 101] | [Name: John] | [Prime_Member: Yes] | [Join_Date: 2023-01-15]
●​ Because it is structured, an analyst can instantly query: "Count all users where
Prime_Member = Yes."

👉 Question Number: 12
👉 Unit Number: 2
Question: Differentiate Spatial data from non Spatial data

👉 Answer:
Definition:

●​ Spatial Data: Also known as geospatial data, it is information that is physically linked to
a specific geographical location on Earth (requires coordinates).
●​ Non-Spatial Data: Also known as attribute data, it is information that describes
characteristics, traits, or qualities without any geographical or location-based context.

Explanation (Data Mapping Flow):

●​ Spatial Flow: GPS Sensor ➔ Collects Latitude/Longitude ➔ Mapped on a Grid ➔ Used

for Routing/Maps
●​ Non-Spatial Flow: User Input ➔ Text/Numbers Collected ➔ Stored in Database ➔
Used for Statistical Analytics

Key Differences:

Feature Spatial Data Non-Spatial Data

1. Core Element Geometry (Coordinates, Latitude, Alphanumeric (Text, Numbers,

Longitude). Dates).

2. Storage Format Vectors (Points, Lines, Polygons) or Standard Relational Database

Rasters (Pixels). Tables.

3. Question "Where is it located?" "What is it?" or "How much is

Answered it?"
 4. Processing Tool GIS (Geographic Information RDBMS, Excel, Standard
Systems). Analytics Tools.

Real-World Example (Amazon):

Imagine an Amazon delivery truck on its route.

●​ Spatial Data: The live GPS coordinates (Lat: 40.7128° N, Long: 74.0060° W) showing
exactly where the truck is currently driving.
●​ Non-Spatial Data: The data detailing that the truck is a "Ford Transit", colored "Blue",
carrying "150 packages" weighing "500 lbs".

👉 Question Number: 13
👉 Unit Number: 3
Question: Differentiate correlation and regression

👉 Answer:
Definition:

According to Geeks for Geeks, Correlation is a statistical technique used to measure the
strength and direction of the linear relationship between two variables. Regression is a
statistical measurement used to determine the exact equation of that relationship, allowing
analysts to predict the value of a dependent variable based on an independent variable.

Explanation (Workflow & Differences):

●​ The Statistical Flow: Identify 2 Variables ➔ Check Correlation (Are they related?) ➔
Apply Regression (Predict exact future values based on the relationship)

Feature Correlation Regression

1. Primary Objective To find out if and how To estimate or predict the

strongly two variables are exact value of one variable
related. based on another.
2. Causation Does not imply causation Implies a cause-and-effect
(just shows they move relationship (X influences Y).
together).

3. Equation/Output A single correlation A mathematical equation (e.g.,

coefficient ($r$) ranging from $Y = mX + c$).
-1 to +1.

4. Variable Roles No difference between Strictly defines an Independent

independent and dependent (Predictor) and Dependent
variables. (Outcome) variable.

5. Visual Representation Represented as a scatter plot Represented as a "Line of Best

of scattered data points. Fit" drawn through the data
points.

Real-World Example (Amazon):

●​ Correlation: An Amazon analyst discovers a strong positive correlation (+0.85) between

"Marketing Ad Spend" and "Prime Video Subscriptions" (they both go up at the same
time).
●​ Regression: The analyst uses regression to predict that for every extra $1,000 spent on
marketing ($X$), Amazon will gain exactly 50 new Prime subscriptions ($Y$).
👉 Question Number: 14
👉 Unit Number: 3
Question: Explain in detail about continuous random variables and its mathematical
equations

👉 Answer:
Definition:
As referenced in Geeks for Geeks, a Continuous Random Variable is a statistical variable that
can take on an infinite number of possible values within a given continuous range or interval.
Unlike discrete variables (which are counted), continuous variables are always measured.

Explanation (Core Concepts):

●​ The Measurement Flow: Identify Variable ➔ Measure (Time/Weight/Distance) ➔ Map to

Continuous Range ➔ Model with Probability Density Function (PDF)
●​ Infinite Possibilities: Between any two values of a continuous variable, there are infinitely
many other valid values (e.g., a weight can be 5.1 kg, 5.12 kg, 5.125 kg, etc.).
●​ Zero Exact Probability: The probability of a continuous random variable taking one exact
specific number is strictly zero. We can only measure the probability of the variable falling
within a range of numbers.

Mathematical Equations:

Continuous random variables are described by a Probability Density Function (PDF), denoted as ​
Non-negativity condition: The probability density must be zero or positive.

1.​ Total Probability: The total area under the probability density curve must exactly equal 1
(representing 100% probability).​

Probability of an Interval: The probability that the variable X falls between two values (a
and b) is the integral (area under the curve) between those points.

PART 6

👉 Question Number: 16
👉 Unit Number: 4
Question: Define data visualization and it’s methods

👉 Answer:
Definition:

As referenced in Geeks for Geeks, Data Visualization is the graphical representation of raw
information and data. By using visual elements like charts, graphs, and maps, data visualization
tools provide an accessible way to see and understand trends, outliers, and patterns in data.

Explanation (The Visualization Flow):

To easily remember how visualization works, think of this flowchart:

Raw Text Data ➔ Choose Visual Method ➔ Map Data to Shapes/Colors ➔ Discover Hidden
Trends ➔ Make Business Decisions

Key Methods of Data Visualization:

There are three primary methods used by data professionals, depending on their goal:

●​ 1. Exploratory Visualization (Finding the Story): Used internally by data analysts. The
goal is to explore massive datasets without a specific hypothesis to see if any unknown
patterns or anomalies exist.
●​ 2. Explanatory/Declarative Visualization (Telling the Story): Used when presenting to
stakeholders or executives. The analyst already knows the answer and uses clean,
simple visuals to explain that specific finding to a non-technical audience.
●​ 3. Interactive Visualization (Exploring the Story): Dashboards where the user can
actively filter, zoom, and drill down into the data themselves to answer their own specific
questions.

Real-World Example (Amazon):

●​ Exploratory: An Amazon Data Scientist creates a massive 3D scatter plot of user clicks
to see if they can find any weird patterns in how people search for products.
●​ Explanatory: The analyst finds that users click "Next Page" less on mobile. They make
a simple Bar Chart showing "Mobile Clicks vs. Desktop Clicks" to present to the CEO.
●​ Interactive: Amazon vendors are given an Interactive Dashboard where they can filter
their product sales by date, region, or device type.

👉 Question Number: 17
👉 Unit Number: 4
Question: Explain in detail about different types of charts and graphs

👉 Answer:
Definition:

Charts and graphs are specific visual tools used in data analytics to represent relationships,
comparisons, distributions, and compositions within a dataset. According to Geeks for Geeks,
selecting the right chart is entirely dependent on the specific business question being asked.

Explanation (Chart Selection Flowchart):

Comparing Categories ➔ Bar Chart

Showing Trends Over Time ➔ Line Chart

Showing Proportions/Percentages ➔ Pie or Donut Chart

Showing Relationships/Correlations ➔ Scatter Plot

Detailed Types of Charts:

●​ Bar/Column Charts: Best for comparing discrete, categorical data. (e.g., Sales of
Laptops vs. Phones).
●​ Line Charts: Best for displaying continuous data over a period of time to show trends,
accelerations, or decelerations.
●​ Pie/Donut Charts: Best for showing "part-to-whole" relationships. They show how a
total amount is divided into proportional slices.
●​ Scatter Plots: Best for showing the correlation between two numeric variables. Data
points are plotted on an X and Y axis to see if they cluster or form a line.
●​ Histograms: Best for showing the distribution of a single continuous variable (e.g.,
grouping customers by age ranges: 10-20, 20-30, 30-40).

Real-World Example (Amazon):

If Amazon wants to visualize its Prime Day performance:

●​ Line Chart: Shows the total number of Prime subscriptions purchased per hour from
12:00 AM to 11:59 PM.
●​ Pie Chart: Shows the percentage breakdown of device usage (60% Mobile App, 30%
Desktop Browser, 10% Smart TV).
●​ Bar Chart: Compares total revenue generated in "North America" vs. "Europe" vs.
"Asia".

👉 Question Number: 18
👉 Unit Number: 4
Question: What do you mean by block charts and what are the significance when
compared to other charts

👉 Answer:
Definition:
A Block Chart (most commonly known in Geeks for Geeks data viz tutorials as a Treemap or
Block Diagram) is a method for displaying complex, hierarchical data using a set of nested
rectangles (blocks). The size and color of each block are directly proportional to a specific data
value.

Explanation (The Block Chart Flow):

Total Entire Dataset (One Giant Box) ➔ Split into Major Categories (Large Blocks) ➔ Split into
Sub-Categories (Smaller Blocks inside Large Blocks) ➔ Color-Coded by Performance

Significance Compared to Other Charts:

●​ 1. Extreme Space Efficiency: While a Bar Chart might get messy and unreadable if you
have 500 categories, a Block Chart can easily fit thousands of tiny blocks onto a single
laptop screen, making the best use of spatial area.
●​ 2. Superior Hierarchical View: Unlike a Pie Chart (which can only show one level of
data), a Block Chart shows "parts-of-a-whole" across multiple levels. You can see the
whole parent category and all its "children" inside it simultaneously.
●​ 3. Dual-Variable Visualization: It allows you to visualize two metrics at once. The Size
of the block represents one metric (like Total Sales), while the Color represents another
(like Profit Margin: Green = Good, Red = Bad).

Real-World Example (Amazon):

Amazon's executive team uses a massive Block Chart to view their entire global revenue.

●​ The entire screen is one giant square representing Amazon's Total Revenue.
●​ It is divided into two main massive blocks: "Amazon E-Commerce" and "Amazon Web
Services (AWS)".
●​ Inside the "E-Commerce" block, there are hundreds of smaller blocks representing
departments ("Electronics", "Clothing", "Groceries").
●​ The size of the "Electronics" block shows it brings in the most money. However, the
block is colored Red, instantly warning executives that while revenue is high, the profit
margin is currently dropping compared to last year.