Scribd
Upload
2 views3 pages

Program4 S Algorithm Explaination

The document outlines the implementation of the Find-S algorithm to derive hypotheses from a set of training data stored in a CSV file. It describes the process of reading the data, initializing the hypothesis, and updating it based on positive examples while ignoring negative ones. The final output is a generalized hypothesis that reflects the consistent attributes of the positive training examples.

Uploaded by

tulasi
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
2 views3 pages

Program4 S Algorithm Explaination

The document outlines the implementation of the Find-S algorithm to derive hypotheses from a set of training data stored in a CSV file. It describes the process of reading the data, initializing the hypothesis, and updating it based on positive examples while ignoring negative ones. The final output is a generalized hypothesis that reflects the consistent attributes of the positive training examples.

Uploaded by

tulasi
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

4. For a given set of training data examples stored in a .

CSV file, implement and demonstrate the Find-S


algorithm to output a description of the set of all hypotheses consistent with the training examples.
import pandas as pd

def find_s_algorithm(file_path):

We are creating a function named find_s_algorithm.

It takes file_path as input (the CSV file containing training data).

data = pd.read_csv(file_path)
We use pandas (pd) to read the dataset.

The CSV file is loaded into a table format called a DataFrame.

Now data contains all training examples.

print("Training data:")
print(data)

 Displays the dataset so we can see the input examples.

attributes = [Link][:-1]
class_label = [Link][-1]

 [Link] gives all column names.


 [:-1] → selects all columns except the last one (these are input attributes).
 [-1] → selects the last column (this is the target/class label).

Example:

Sky Temp Humidity Wind Play

 Attributes → Sky, Temp, Humidity, Wind


 Class label → Play

hypothesis = ['?' for _ in attributes]

 We create an initial hypothesis.


 ? means most general value.
 If there are 4 attributes → hypothesis = ['?', '?', '?', '?']

This means:

"We don’t know anything yet."

for index, row in [Link]():


 iterrows() goes row by row.
 row contains one training example at a time.

if row[class_label] == 'Yes':

 Find-S works only on positive examples.


 If class label is Yes, we update hypothesis.
 If class label is No, we ignore it.

for i, value in enumerate(row[attributes]):


We compare each attribute value with the current hypothesis.
if hypothesis[i] == '?' or hypothesis[i] == value:
hypothesis[i] = value
else:
hypothesis[i] = '?'

Three cases:

Case 1️⃣: Hypothesis is '?'

 Replace it with the current value.

Example:

 Hypothesis: ['?', '?', '?']


 First positive example: Sunny, Warm, Normal
 New hypothesis: ['Sunny', 'Warm', 'Normal']

Case 2️⃣: Same Value

 Keep it as it is.

Example:

 Hypothesis: ['Sunny', 'Warm', 'Normal']


 Next example: Sunny, Warm, High
 First two match → keep them.

Case 3️⃣: Different Value

 Replace with ?

Example:

 Hypothesis: ['Sunny', 'Warm', 'Normal']


 Next example: Rainy, Warm, Normal
 First attribute different → change to ?
 New hypothesis: ['?', 'Warm', 'Normal']
return hypothesis
After checking all positive examples, we return the final hypothesis.
file_path = 'training_data.csv'
hypothesis = find_s_algorithm(file_path)

 We give the CSV file path.

 The function runs and calculates the hypothesis.

print("\nThe final hypothesis is:", hypothesis)

 Displays the final generalized rule.

You might also like