ass-12(Hiararchichal )

 import pandas as pd 

import numpy as np 

import matplotlib.pyplot as plt 

import seaborn as sns 

 

from sklearn.preprocessing import StandardScaler 

from scipy.cluster.hierarchy import dendrogram, linkage 

from sklearn.cluster import AgglomerativeClustering 

# Load dataset 

df = pd.read_csv('Vehicle_dataset.csv') 

 

# Step 1: Load & Clean Data 

df.replace('$null$', np.nan, inplace=True) 

df.dropna(inplace=True) 

 

# Convert necessary columns to numeric 

cols_to_convert = ['engine_s', 'horsepow', 'wheelbas', 'width', 'length', 

                   'curb_wgt', 'fuel_cap', 'mpg'] 

df[cols_to_convert] = df[cols_to_convert].astype(float) 

 

# Step 2: Select Features for Clustering 

features = df[cols_to_convert] 

# Step 3: Standardize Features 

scaler = StandardScaler() 

scaled_features = scaler.fit_transform(features) 

 

# Step 4 & 5: Dendrograms with 3 Linkage Types 

linkage_methods = ['single', 'complete', 'average'] 

labels = df['manufact'] + ' ' + df['model'] 

 

for method in linkage_methods: 

    plt.figure(figsize=(12, 5)) 

    Z = linkage(scaled_features, method=method) 

    dendrogram(Z, labels=labels.values, leaf_rotation=90) 

    plt.title(f'Dendrogram - {method.capitalize()} Linkage') 

    plt.xlabel('Car Model') 

plt.ylabel('Distance') 

plt.tight_layout() 

plt.grid(True) 

plt.show()