Job Saarnee


Top Most Platform for Job Updates O Level Results Answer keys AKTU MCQs

Find S Algorithm with Example in Machine Learning | Machine Learning

Find S Algorithm in Machine Learning


In Find S algorithm we tend to find a Maximally Specific Hypothesis that fits the all positive training examples.

Introduction to Find S Algorithm:

“Welcome to our in-depth exploration of the groundbreaking Find-S Machine Learning Algorithm! In this comprehensive blog, we unravel the complexities of Find-S, offering insights into its applications, benefits, and real-world impact. Whether you’re a data enthusiast, a tech professional, or simply curious about the future of machine learning, our detailed analysis will guide you through the innovative realm of Find-S. Join us as we demystify this powerful algorithm and discover how it’s reshaping the landscape of artificial intelligence.”

Some important points related to Find –S algorithm:

  1. Find-S algorithm only considers positive training examples and neglect negative training examples.
  2. In Find-S algorithm, we move from top to bottom i.e. specific hypothesis to general hypothesis. In the other words we can say that in Find-S algorithm we start with the most specific hypothesis and generalizes this hypothesis each time whenever the attributes values of hypothesis and attributes values of observed positive training example did not match.
  3. Maximally specific hypothesis A hypothesis h, is a maximally specific hypothesis if it covers none of the negative training examples and there is no other hypothesis h’ that covers none negative training examples, such that h is strictly more general than h’.

Find S Algorithm

Notations used in Find-s algorithm:

  1. The most specific hypothesis is represented by the by the {φ,φ,φ,φ} where number of the ‘φ’ is equal to number of attributes in training data.
  2. ‘φ’ indicate that no value is acceptable for the attributes.
  3. ‘?’ Indicate that any value can be acceptable for the attributes.

Find –s algorithm:

Step 1: Initialize h to most specific hypothesis h.
Step 2: For each positive training instance x
Step 3: for each attribute’s constraint ai in h
if the constraint ai is satisfied by x Then does nothing
replace ai in h by the next general hypotheses Constraint ‘?’ that is satisfied by x.
Step 4: Output hypothesis.

Example: – To understand this algorithm, we consider the below training example.





Play tennis





















In the above training example, target attributes are play Tennis.

First, we initialize h to most specific hypothesis:

h = {φ, φ, φ, φ}

Now we consider first training example:

x1 = (Overcast, Hot, High, Weak)

This is the positive training example.  From here, it is clear that none of the attributes value in h is satisfied with the attributes value in x1.

So, each attribute in h is replaced by the next general constraints –

h1 = (Overcast, Hot, High, Weak)

Now, we consider second training example:

x2 = (Rain, Mild, High, Weak)

This is positive training example.

We compare each attribute value in h1 with the attributes value in x2 and substitute ‘?’ in the place of any attributes value in h if it is not satisfied with x2.

h2 = (‘?’, ‘?’, High, Weak)

Now, we consider third training example:

x3 = (Rain, Cold, Normal, Normal)

This is negative training example. so, we neglect this training example and proceed further.

Now we consider fourth training example.

x4 = (Overcast, Cool, Normal, Weak)

This is positive training example.

After comprising each attribute value in h2 with the attributes value in x4,

we have-

h3 = (‘?’, ‘?’, ‘?’, weak)

This is the output hypothesis.

For More technical topic please visit our another site 

Machine Learning 99+ Most Important MCQ

Leave a Comment

Your email address will not be published. Required fields are marked *

Shopping Cart

You cannot copy content of this page