Class 10 AI Chapter - Modelling in AI Topic - Machine Learning - Arvindzeclass

Class 10 Artificial Intelligent Code 417 Solutions

Session 2025-26

Artificial Intelligence code 417 syllabus pdf class 10 solutions. Class 10 AI Book. Class 10 AI Notes. This article provides complete solution for class 10 AI (Artificial Intelligence) Code - 417 students according to new syllabus 2025 – 26. In this article first of all go through the AI Code - 417 syllabus and then follow chapter wise notes.

Part A - Employability Skills Notes

Chapter 1 - Communication Skills - II

Chapter 2 - Self - Management Skills - II

Chapter 3 - ICT Skill - II

Chapter 4 - Entrepreneurial Skills - II

Chapter 5 - Green Skills - II

Part B - Subject Specific Skills Notes

Chapter 1 - AI Project Cycle & Ethics

Chapter 2 - Modelling in AI

Chapter 3 - Evaluating Models

Chapter 4 -Statistical Data

Chapter 5 -Computer Vision

Chapter 6 - Natural Language Processing

Chapter 7 - Advance Python

Supervised Learning

Supervised Learning is a machine learning technique where:

The machine is given input data along with the correct output.
The goal is to learn a mapping from inputs to outputs.
After training, the model can predict the output for new, unseen inputs.

Supervised Learning

Think of it like:

A teacher giving a student (AI) many practice problems with correct answers. The student learns the pattern, and later solves similar problems without help.

Example:

Imagine you're training a model to recognize animals:

Input: Images of animals
Output: Labels like "Cat", "Dog", "Elephant"

The model learns which features (like ears, tail, size) match with which animal.
Later, when given a new image, it tries to predict the correct animal.

AI Model

Subsets of Supervised Learning

There are two main types based on the kind of output the model predicts:

1. Classification

Definition:

Classification is used when the output is a category or label.

Purpose:

To assign each input to one of a set of predefined classes.

Examples:

Email: Spam or Not Spam
Medical: Disease Present or Not
Image: Cat, Dog, or Bird
Exam: Pass or Fail

Output:

Discrete — e.g., {0, 1}, {red, green, blue}, or {cat, dog, horse}

Algorithms used:

Decision Trees
Logistic Regression
Support Vector Machines (SVM)
Naive Bayes
K-Nearest Neighbors (KNN)

2. Regression

Definition:

Regression is used when the output is a real number or continuous value.

Purpose:

To predict a quantity based on input features.

Examples:

Predict house prices based on size, location, etc.
Forecast temperature for the next week.
Estimate a person's weight based on age and height.
Predict number of customers in a store.

Output:

Continuous — e.g., 45.2, 1050.75, 0.89

Algorithms used:

Linear Regression
Polynomial Regression
Support Vector Regression (SVR)
Decision Tree Regression

Key Differences at a Glance:

Feature	Classification	Regression
Output Type	Categories or labels	Continuous numeric values
Example Task	Identify an email as spam or not	Predict the price of a house
Nature of Output	Discrete	Continuous
Common Algorithms	Logistic Regression, SVM, KNN	Linear Regression, SVR
Evaluation Metrics	Accuracy, F1-score, Confusion Matrix	MSE, RMSE, R² (R-squared)

Unsupervised Learning

Unsupervised learning is a type of machine learning where:

The system is given only input data (no labels or correct answers).
It learns to identify patterns, structures, or relationships within the data without supervision.

📌 In simple terms:
The AI is like an explorer without a map — it groups or organizes information on its own, without being told what's right or wrong.

Real-Life Analogy:

Imagine you have a basket of mixed fruits, but none are labeled. You start sorting them by shape, color, or size. That’s what unsupervised learning does — it finds natural groupings without being told the category names.

Unsupervised Learning

Subcategories of Unsupervised Learning

1. Clustering

Definition:

Clustering is the process of grouping similar items based on their features.

Purpose:

To divide the dataset into meaningful groups (clusters) where items in the same group are more similar to each other than to those in other groups.

Examples:

Grouping customers by purchasing habits
Segmenting students based on learning style
Grouping news articles by topic

Common Algorithms:

K-Means Clustering
Hierarchical Clustering
DBSCAN (Density-Based Spatial Clustering)

Output:

Several clusters (e.g., Cluster 1, Cluster 2, etc.), without any predefined labels.

2. Association

Definition:

Association learning finds relationships or patterns between items in large datasets.

Purpose:

To discover interesting rules or associations among variables in the data.

Examples:

Market Basket Analysis (e.g., “If a customer buys bread, they are likely to buy butter”)
Recommendation systems (e.g., YouTube suggesting videos)
Web usage mining (e.g., users who visit page A also visit page B)

Common Algorithms:

Apriori Algorithm
Eclat Algorithm
FP-Growth Algorithm

Output:

Association rules, usually in the format:

If X, then Y (with a certain confidence and support)

If a customer buys Milk and Bread, then they are likely to buy Butter.

Comparison Table:

Feature	Clustering	Association
Purpose	Group similar data points	Find relationships between variables
Output	Groups or clusters	If-then rules
Data Required	Unlabeled	Unlabeled
Example	Grouping customers	Market basket analysis
Algorithm Example	K-Means, DBSCAN	Apriori, FP-Growth

Reinforcement Learning

Reinforcement Learning (RL) is a type of Machine Learning in Artificial Intelligence (AI) where an agent learns to make sequences of decisions by interacting with an environment to achieve a goal.

The agent receives feedback in the form of rewards or penalties based on its actions and aims to maximize the total cumulative reward over time.

Basic Elements of Reinforcement Learning:

Component	Description
Agent	The learner or decision-maker.
Environment	The system with which the agent interacts.
State (S)	The current situation of the environment.
Action (A)	Choices the agent can make.
Reward (R)	Numerical value received after an action (positive or negative).
Policy (π)	Strategy used by the agent to choose actions.
Value Function	Predicts future reward from a state.

Reinforcement Learning Process:

Agent observes the current state.
Chooses an action based on a policy.
The environment returns a reward and new state.
The agent updates its policy using the reward.
Repeat until the agent learns the optimal strategy.

Reinforcement Learning

Applications of Reinforcement Learning:

1. Game Playing

Example:

AlphaGo (by DeepMind), Dota 2 AI, Chess, and Atari Games.

How it works:

The agent (AI) plays games against itself or others.
It learns which moves lead to wins (positive reward) or losses (negative reward).
Over time, it develops strategies better than human experts.

2. Self-Driving Cars

What RL does:

Teaches the car how to drive safely in real-world conditions.

Learning tasks:

Lane following
Obstacle avoidance
Decision-making at intersections

Reward system:

Positive reward for staying in lane or reaching destination.
Penalties for accidents, hard braking, or disobeying traffic rules.

3. Robotics

Use:

Robotic arms, biped robots, drones.

Examples:

A robot learns to walk by receiving feedback on balance.
A robotic hand learns to pick and place objects.

Learning Process:

The robot tries actions (e.g., moving joints), observes outcomes (falls or balances), and learns better control.

4. Healthcare and Medicine

Use Case:

Personalized treatment plans, medical diagnosis.

How:

RL is used to optimize treatment over time.
For example, choosing the correct drug dosage based on patient response.

Benefits:

Minimizes side effects.
Improves treatment outcomes.

5. Recommendation Systems

Example:

YouTube, Netflix, Amazon.

What RL does:

Learns which content keeps the user engaged.
Rewards based on user actions like clicking, liking, or watching time.

Result:

Dynamically adjusts content recommendations to maximize engagement.

6. Finance and Trading

How it’s used:

RL agents learn to buy/sell stocks or cryptocurrencies.
They optimize for long-term profit.

Features:

Observes market data (state).
Takes actions (buy, sell, hold).
Gets rewards based on profit/loss.

7. Internet of Things (IoT) & Smart Systems

Example:

Smart HVAC systems, energy optimization in smart homes.

Use:

RL adjusts heating/cooling based on usage patterns.
Learns user behavior and reduces energy consumption.

Summary Table of RL Applications:

Domain	Application	Goal of RL
Gaming	Chess, Go, Atari,	Win the game or reach high score
Robotics	Grasping, walking, flying	Improve control and movement
Self-Driving	Navigation, traffic decisions	Safe and efficient driving
Healthcare	Treatment strategies	Optimize patient health outcomes
Recommendation	YouTube, Amazon, Netflix	Maximize user engagement and satisfaction
Finance	Trading, portfolio management	Maximize return and reduce risk
Smart Devices	Energy saving, smart thermostats	Efficient system operation with minimal cost