Last Updated on April 20, 2026 by Rajeev Bagra

When beginners first see this code:

class Node():
    def __init__(self, state, parent, action):
        self.state = state
        self.parent = parent
        self.action = action

…it may look small and simple.

But this tiny class powers some of the most important ideas in Artificial Intelligence, pathfinding, graphs, game logic, and problem solving systems.

It helps a computer answer questions like:

• How do I solve this maze?
• What is the shortest route?
• How did I reach this answer?
• What move should I make next?

---

What Is a Node?

A Node represents one step in a journey.

Think of it like a checkpoint containing memory.

Each node stores three things:

Attribute	Meaning
state	Current position or situation
parent	Previous node you came from
action	Move taken to get here

---

Visual Representation

[Current Node]
 State  = Mumbai
 Parent = Pune
 Action = Drive

Meaning:

I am currently in Mumbai.
I came from Pune.
I got here by driving.

---

Why This Is Powerful

A computer often explores many possibilities.

Without memory, it may know only where it is now.

With Node, it remembers the path.

---

Journey Example

Suppose you travel:

Pune → Mumbai → Surat → Ahmedabad

This can be stored like:

Ahmedabad
   ↑
Surat
   ↑
Mumbai
   ↑
Pune

Each city is linked to the previous one through parent.

---

The Code Explained

class Node():

Creates a blueprint for making nodes.

---

def __init__(self, state, parent, action):

Runs automatically when a new node is created.

---

self.state = state

Stores where you are now.

Examples:

• A city
• A chess board position
• A webpage
• A maze location

---

self.parent = parent

Stores the previous node.

This creates a chain.

---

self.action = action

Stores what move got you here.

Examples:

• Move left
• Drive
• Click link
• Jump knight

---

Graphic: How Parent Links Work

Goal Node
   |
   v
[Ahmedabad]
   |
 Parent
   |
   v
[Surat]
   |
 Parent
   |
   v
[Mumbai]
   |
 Parent
   |
   v
[Pune]

When goal is found, the program follows parents backward.

---

Real Use Cases

---

1. Maze Solving

Start → Right → Down → Left → Exit

Each move becomes a node.

When exit is found:

The program reconstructs the path.

---

2. Google Maps Style Routing

Pune → Mumbai → Nashik → Delhi

Every city can be a node.

Used to compute routes.

---

3. Chess & Games

Each board position = node.

Parent = previous move.

Action = move made.

Used in game AI.

---

4. Website Navigation

Find shortest clicks from homepage to checkout.

Each webpage = node.

---

Example in Python

start = Node("Pune", None, None)

mumbai = Node("Mumbai", start, "Drive")

surat = Node("Surat", mumbai, "Drive")

ahmedabad = Node("Ahmedabad", surat, "Drive")

---

Graphic of Above

[Pune]
   ↓ Drive
[Mumbai]
   ↓ Drive
[Surat]
   ↓ Drive
[Ahmedabad]

---

Why parent = None at Start?

The first node has no previous step.

So:

start = Node("Pune", None, None)

Means:

• Starting point
• No parent
• No action needed yet

---

Why This Matters in AI

This class is heavily used in:

• Breadth First Search (BFS)
• Depth First Search (DFS)
• A* Search
• Route Planning
• Robotics
• Puzzle Solvers

Without nodes, search becomes messy.

With nodes, search becomes organized.

---

Deep Insight

A node models this truth:

Every current state came from a previous decision.

That idea is central to planning and intelligence.

---

Beginner-Friendly Analogy

Imagine breadcrumbs in a forest.

Each breadcrumb says:

• You are here
• You came from there
• You walked this way

That is exactly what a node does.

---

Final Summary

class Node():
    def __init__(self, state, parent, action):
        self.state = state
        self.parent = parent
        self.action = action

This tiny class helps computers remember journeys, reconstruct paths, and solve complex problems intelligently.

Small code.

Massive impact.

---

Key Takeaway Graphic

Node = Current State + Previous Step + Action Taken

---

Want to Learn Next?

A perfect next topic would be:

StackFrontier vs QueueFrontier

These two classes decide how nodes are explored, and they power DFS vs BFS.