10/20/24, 11:27 PM ml-exp-5 (1) - Jupyter Notebook

In [5]:  import pandas as pd

import numpy as np
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler
​
# Load the dataset with a different encoding
url = '/kaggle/input/sample-sales-data/sales_data_sample.csv'
df = pd.read_csv(url, encoding='ISO-8859-1') # Adjust encoding if need
​
# Display the first few rows of the dataset
print(df.head())
​
# Check for missing values
print("\nMissing values in each column:")
print(df.isnull().sum())
​
​

localhost:8888/notebooks/Downloads/ml-exp-5 (1).ipynb 2/7

10/20/24, 11:27 PM ml-exp-5 (1) - Jupyter Notebook

ORDERNUMBER QUANTITYORDERED PRICEEACH ORDERLINENUMBER SALES

\
0 10107 30 95.70 2 2871.00
1 10121 34 81.35 5 2765.90
2 10134 41 94.74 2 3884.34
3 10145 45 83.26 6 3746.70
4 10159 49 100.00 14 5205.27

ORDERDATE STATUS QTR_ID MONTH_ID YEAR_ID ... \

0 2/24/2003 0:00 Shipped 1 2 2003 ...
1 5/7/2003 0:00 Shipped 2 5 2003 ...
2 7/1/2003 0:00 Shipped 3 7 2003 ...
3 8/25/2003 0:00 Shipped 3 8 2003 ...
4 10/10/2003 0:00 Shipped 4 10 2003 ...

ADDRESSLINE1 ADDRESSLINE2 CITY STATE

\
0 897 Long Airport Avenue NaN NYC NY
1 59 rue de l'Abbaye NaN Reims NaN
2 27 rue du Colonel Pierre Avia NaN Paris NaN
3 78934 Hillside Dr. NaN Pasadena CA
4 7734 Strong St. NaN San Francisco CA

POSTALCODE COUNTRY TERRITORY CONTACTLASTNAME CONTACTFIRSTNAME DEALS

IZE
0 10022 USA NaN Yu Kwai Sm
all
1 51100 France EMEA Henriot Paul Sm
all
2 75508 France EMEA Da Cunha Daniel Med
ium
3 90003 USA NaN Young Julie Med
ium
4 NaN USA NaN Brown Julie Med
ium

[5 rows x 25 columns]

Missing values in each column:

ORDERNUMBER 0
QUANTITYORDERED 0
PRICEEACH 0
ORDERLINENUMBER 0
SALES 0
ORDERDATE 0
STATUS 0
QTR_ID 0
MONTH_ID 0
YEAR_ID 0
PRODUCTLINE 0
MSRP 0
PRODUCTCODE 0
CUSTOMERNAME 0
PHONE 0
ADDRESSLINE1 0
ADDRESSLINE2 2521
CITY 0
STATE 1486
POSTALCODE 76
COUNTRY 0
TERRITORY 1074
localhost:8888/notebooks/Downloads/ml-exp-5 (1).ipynb 3/7
10/20/24, 11:27 PM ml-exp-5 (1) - Jupyter Notebook
CONTACTLASTNAME 0
CONTACTFIRSTNAME 0
DEALSIZE 0
dtype: int64

In [6]:  # Drop columns that are not needed for clustering

# For demonstration, we'll only use relevant numerical columns for clus
df_cleaned = df[['QUANTITYORDERED', 'PRICEEACH']]
​
# Handle missing values (if necessary, e.g., dropping or filling)
df_cleaned = df_cleaned.dropna()
​
# Scale the data
scaler = StandardScaler()
scaled_data = scaler.fit_transform(df_cleaned)
​
​

localhost:8888/notebooks/Downloads/ml-exp-5 (1).ipynb 4/7

10/20/24, 11:27 PM ml-exp-5 (1) - Jupyter Notebook

In [7]:  # Determine the number of clusters using the elbow method

inertia = []
K = range(1, 11)
for k in K:
kmeans = KMeans(n_clusters=k, random_state=42)
kmeans.fit(scaled_data)
inertia.append(kmeans.inertia_)
​
# Plot the elbow curve
plt.figure(figsize=(10, 6))
plt.plot(K, inertia, 'bo-')
plt.xlabel('Number of clusters (k)')
plt.ylabel('Inertia')
plt.title('Elbow Method for Optimal k')
plt.xticks(K)
plt.grid()
plt.show()
​

localhost:8888/notebooks/Downloads/ml-exp-5 (1).ipynb 5/7

10/20/24, 11:27 PM ml-exp-5 (1) - Jupyter Notebook

localhost:8888/notebooks/Downloads/ml-exp-5 (1).ipynb 7/7