Chapter 23: Data Exploration
Introduction
Data exploration is the process of analyzing and visualizing data to understand its structure,
patterns, and insights. It helps in detecting anomalies, trends, and relationships within datasets.
Data Visualization
Data visualization refers to representing data in graphical formats such as charts, graphs, and
maps. It helps in understanding data trends, outliers, and patterns effectively.
Data Visualization Tools
Some commonly used data visualization tools include:
1. MS Excel
2. Google Sheets
3. Tableau
4. Power BI
5. Matplotlib (Python Library)
Types of Data Visualization
1. Bar Chart – Used to compare different categories.
2. Line Chart – Used to show trends over time.
3. Pie Chart – Used to show proportions.
4. Histogram – Used for distribution of numerical data.
5. Scatter Plot – Used to show relationships between variables.
Short Answer Questions
1. Briefly explain the process of creating 3-D columns in MS Excel.
o In MS Excel, select the data, go to Insert tab, choose Column Chart, and select 3-
D column type.
2. Mention 5 different forms of data visualization.
o Bar Chart, Line Chart, Pie Chart, Histogram, Scatter Plot.
3. What do you know about Google Sheets? How can we use them for data
visualization?
o Google Sheets is an online spreadsheet tool that allows users to create and edit
spreadsheets. It offers chart options like bar, pie, and line charts for visualization.
4. Why is MS Excel one of the most popular tools for data visualization?
o MS Excel provides user-friendly tools, multiple chart options, data analysis
features, and is widely available.
5. What is a combination chart? Explain their usage.
o A combination chart combines two or more chart types (e.g., bar and line) to
represent different datasets on the same graph.
Chapter 24: Data Modeling
Introduction
Data modeling is the process of creating a visual representation of a system's data and its
relationships.
AI, ML, and DL
AI (Artificial Intelligence): Simulates human intelligence.
ML (Machine Learning): AI subset that learns from data.
DL (Deep Learning): ML subset using neural networks.
Modeling
Modeling refers to designing AI models to process and analyze data.
AI Models
Rule-Based Models: Use predefined rules for decision-making.
Learning-Based Models: Learn patterns from data.
Types of Learning-Based Approaches
1. Supervised Learning – Uses labeled data.
2. Semi-Supervised Learning – Uses a mix of labeled and unlabeled data.
3. Unsupervised Learning – Finds patterns in unlabeled data.
4. Reinforcement Learning – Uses rewards to learn optimal actions.
Application of Learning-Based Approaches
Classification: Categorizing data into predefined groups.
Prediction: Forecasting outcomes.
Regression: Predicting continuous values.
Decision Trees: Structured decision-making models.
Dimensionality Reduction: Reducing the number of variables.
Evaluation
AI models are evaluated using accuracy, precision, recall, and F1-score.
Neural Networks
Working of Neural Networks: Layers of interconnected nodes process data.
Perceptron: Basic unit of neural networks.
Short Answer Questions
1. Difference between ML and DL?
o ML needs structured data, while DL uses neural networks.
2. Working principle of rule-based models?
o Uses predefined rules for decision-making.
3. Advantages of learning-based models over rule-based models?
o They adapt to new data and improve over time.
4. Types of nodes in a decision tree?
o Root, internal, and leaf nodes.
5. What is classification in AI?
o Categorizing data into predefined groups.
6. What is regression in AI?
o Predicting continuous values.
7. What is splitting in decision trees?
o Dividing data into subsets.
8. Explain dimensionality reduction with an example.
o Reducing features to simplify models.
9. Describe the operation of a perceptron.
o Processes inputs through weighted connections and activation functions.
10. Parameters to evaluate an AI project?
Accuracy, precision, recall, and F1-score.
Chapter 25: Anaconda and Jupyter Notebook
Introduction
Anaconda is a distribution for Python that simplifies AI coding.
Coding for AI
Python is the primary language for AI development.
Different Modes of Coding in Python
Interactive Mode
Script Mode
Anaconda and Jupyter Notebook
Installing Anaconda
How to Access Anaconda:
o Using Command Prompt
o Using Anaconda Navigator
Virtual Environment in Anaconda
Allows isolated project environments.
Short Answer Questions
1. What is Anaconda? Why do we need it for AI projects?
o It is a package manager and environment for Python.
2. What is Jupyter Notebook?
o An interactive coding environment for Python.
3. Ways to install Anaconda on Windows?
o Via graphical installer or command line.
4. What is a virtual environment in Anaconda?
o A separate environment for managing dependencies.
5. How to save files in Jupyter Notebook?
o Using the File > Save option or shortcut Ctrl+S.
Chapter 26: Python
Introduction
Python is a high-level programming language known for its simplicity.
Features of Python
Easy to learn
Interpreted language
Extensive libraries
Basics of Python
print() function: Displays output.
Comments: # for single-line, ''' for multi-line.
Character Set: Letters, digits, symbols.
Reserved Keywords: Predefined words.
Identifiers: Variable names.
Statements: Instructions in code.
Indentation: Code structure in Python.
Variables: Store data values.
Data Types: int, float, str, bool, etc.
Input and Output
input(): Accepts user input.
print(): Displays output.
Operators
Arithmetic: +, -, *, /
Comparison: ==, !=, >, <
Assignment: =, +=, -=
Logical: and, or, not
Bitwise: &, |, ^
Conditional Statements
if, if-else, if-elif-else
Loops
For loop
While loop
Loop control statements: break, continue
Nested loops