Page 231 - Touhpad Ai

Program 31: To compare students' marks from different subjects using Z-score normalization

import pandas as pd
from sklearn.preprocessing import StandardScaler
# Sample data
data = {'Marks': [65, 75, 80, 92, 100]}
df = pd.DataFrame(data)
# Z-score normalization
scaler = StandardScaler()
df['Z_Score'] = scaler.fit_transform(df[['Marks']])
print(df)
Output:

Marks Z_Score
0 65 -1.406523
1 75 -0.598176
2 80 -0.194003
3 92 0.776013
4 100 1.422690
Scale Adjustment (Unit Conversion)
Scale adjustment ensures all data uses the same units of measurement.

Program 32: To convert height from feet to centimetres or weight from pounds to kilograms
# Sample data
df = pd.DataFrame({'Weight_pounds': [120, 152, 185]})
# Convert to kilograms (1 pound = 0.453592 kg)

df['Weight_kg'] = df['Weight_pounds'] * 0.453592
print(df)
Output:

Weight_pounds Weight_kg
0 120 54.431040
1 152 68.945984
2 185 83.914520
Feature Scaling (Min-Max Normalization)
Feature scaling (Min–Max Normalization) changes values to a fixed range, usually between 0 and 1. It helps in
comparing different types of data equally.

Program 33: to scale salary data between 0 and 1 for analysis
from sklearn.preprocessing import MinMaxScaler

# Sample data
df = pd.DataFrame({'Salary': [20000, 30000, 40000, 50000, 60000]})
# Min-Max Scaling
scaler = MinMaxScaler()

df['Salary_Scaled'] = scaler.fit_transform(df[['Salary']])
print(df)

Theoretical and Practical Aspects of Data Processing 229