What is Markov Chains?

S8-SA2-0179

Grade Level:

Class 8

AI/ML, Data Science, Research, Journalism, Law, any domain requiring critical thinking

Definition

What is it?

A Markov Chain is a mathematical model that describes a sequence of events where the probability of each event depends ONLY on the state of the previous event, not on any events before that. Think of it like a chain reaction where the next step only cares about the current step, not how you got there.

Simple Example

Quick Example

Imagine you are choosing what to eat for lunch every day. If you had biryani yesterday, there's a 70% chance you'll have dal-roti today and a 30% chance you'll have biryani again. If you had dal-roti yesterday, there's an 80% chance you'll have biryani today and a 20% chance you'll have dal-roti again. Your choice today only depends on what you ate yesterday, not on what you ate two days ago.

Worked Example

Step-by-Step

Let's say a student's study habit can be 'Studying' (S) or 'Playing' (P). --- If they are 'Studying' today, there's a 90% chance they will be 'Studying' tomorrow and a 10% chance they will be 'Playing'. --- If they are 'Playing' today, there's a 40% chance they will be 'Studying' tomorrow and a 60% chance they will be 'Playing'. --- QUESTION: If the student is 'Studying' today, what is the probability they will be 'Playing' two days from now? --- STEP 1: Student is 'Studying' (S) today. --- STEP 2: Probability for tomorrow: P(S to S) = 0.9, P(S to P) = 0.1. --- STEP 3: Now, consider two paths to be 'Playing' on day 2: --- Path A: S -> S -> P. Probability = P(S to S) * P(S to P) = 0.9 * 0.1 = 0.09. --- Path B: S -> P -> P. Probability = P(S to P) * P(P to P) = 0.1 * 0.6 = 0.06. --- STEP 4: Add probabilities of all paths leading to 'Playing' on day 2. Total Probability = 0.09 + 0.06 = 0.15. --- ANSWER: The probability the student will be 'Playing' two days from now is 15%.

Why It Matters

Markov Chains are super useful in predicting future events based on current situations, making them a core part of AI and Machine Learning. They help in fields like weather forecasting, understanding customer behaviour in business, and even predicting stock market trends. Data Scientists and AI Engineers use this concept daily.

Common Mistakes

MISTAKE: Thinking the future state depends on the entire past history. | CORRECTION: Remember, in a Markov Chain, the future state only depends on the *current* state, not how you got there. This is called the 'memoryless property'.

MISTAKE: Confusing probabilities from different states. | CORRECTION: Always make sure you're using the correct 'transition probabilities' – the chances of moving from *one specific state* to another specific state.

MISTAKE: Not considering all possible paths when calculating probabilities over multiple steps. | CORRECTION: For multi-step predictions, draw a small diagram or list all possible sequences of states that lead to your desired outcome, then sum their probabilities.

Practice Questions

Try It Yourself

QUESTION: A traffic light is either Green (G) or Red (R). If it's Green, there's an 80% chance it stays Green and 20% chance it turns Red. If it's Red, there's a 30% chance it turns Green and 70% chance it stays Red. If the light is Green now, what is the probability it will be Red next? | ANSWER: 20%

QUESTION: Using the traffic light example from Q1, if the light is Red now, what is the probability it will be Green two steps from now? (Hint: Consider R->G->G and R->R->G paths) | ANSWER: 0.3 * 0.8 + 0.7 * 0.3 = 0.24 + 0.21 = 0.45 or 45%

QUESTION: A shop sells two types of sweets: Laddoo (L) and Gulab Jamun (G). If a customer buys Laddoo, the next customer has a 60% chance of buying Laddoo and 40% of buying Gulab Jamun. If a customer buys Gulab Jamun, the next customer has a 20% chance of buying Laddoo and 80% of buying Gulab Jamun. If the first customer bought Laddoo, what is the probability that the third customer will buy Gulab Jamun? | ANSWER: (0.6 * 0.4) + (0.4 * 0.8) = 0.24 + 0.32 = 0.56 or 56%

MCQ

Quick Quiz

Which of the following best describes the core idea of a Markov Chain?

The future depends on all past events.

The future depends only on the present state.

Events happen randomly without any pattern.

The past can be perfectly predicted from the future.

The Correct Answer Is:

A Markov Chain's key feature is its 'memoryless' property, meaning the probability of the next state depends only on the current state, not on the sequence of events that led to it. Options A, C, and D contradict this fundamental idea.

Real World Connection

In the Real World

Many apps you use daily leverage Markov Chains! For example, when your phone's keyboard suggests the next word as you type, it often uses a Markov Chain to predict which word is most likely to follow the current one. Similarly, Google's PageRank algorithm, which decides which webpages appear first in your search results, uses a form of Markov Chains to understand how websites link to each other.

Key Vocabulary

Key Terms

STATE: A specific condition or situation in a sequence (e.g., 'Studying', 'Playing') | TRANSITION: Moving from one state to another | PROBABILITY: The chance or likelihood of an event happening | MEMORYLESS PROPERTY: The future state depends only on the current state, not on past states.

What's Next

What to Learn Next

Great job understanding Markov Chains! Next, explore 'State Transition Diagrams' to visualize these chains, and then 'Steady-State Probabilities' to learn about long-term predictions. These concepts will help you build on your understanding of how systems evolve over time.