source: @allison_horst https://github.com/allisonhorst/penguins
You have been asked to support a team of researchers who have been collecting data about penguins in Antartica! The data is available in csv-Format as penguins.csv
Origin of this data : Data were collected and made available by Dr. Kristen Gorman and the Palmer Station, Antarctica LTER, a member of the Long Term Ecological Research Network.
The dataset consists of 5 columns.
| Column | Description |
|---|---|
| culmen_length_mm | culmen length (mm) |
| culmen_depth_mm | culmen depth (mm) |
| flipper_length_mm | flipper length (mm) |
| body_mass_g | body mass (g) |
| sex | penguin sex |
Unfortunately, they have not been able to record the species of penguin, but they know that there are at least three species that are native to the region: Adelie, Chinstrap, and Gentoo. Your task is to apply your data science skills to help them identify groups in the dataset!
# Import Required Packages
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler
import seaborn as sns
import numpy as np
# Loading and examining the dataset
penguins_df = pd.read_csv("penguins.csv")
penguins_df.head()# penguin types
penguin_types = ['Adelie', 'Chinstrap', 'Gentoo']Only numeric columns
penguins_df.info()penguins_df.describe()penguins_df['sex'].value_counts()sns.pairplot(penguins_df, diag_kind='kde')- We can see initial groupings within the data, for example in the scatterplot of culmen_length_mm vs. flipper_length_mm we can start to see 3 clusters.
- The kde plot for each variable seems fairly normal
Clustering penguins
One-hot encoding is not recommended for k-means clustering
K-means clustering relies on Euclidean distance to measure similarity between points. One-hot encoded categorical variables (e.g., gender_Male = 0 or 1, gender_Female = 0 or 1) create binary vectors, and Euclidean distance between such vectors may not meaningfully represent the similarity between categories.
- If you proceed with one-hot encoding, ensure that all features are scaled (e.g., using standardization or normalization) to avoid bias toward features with larger magnitudes.
# only numeric columns
penguins = pd.get_dummies(penguins_df)
penguins.head()# Normalize data
scaler = StandardScaler()
X_scaled = scaler.fit_transform(penguins)
X_scaledkmeans = KMeans(n_clusters=3, random_state=42)
penguins['cluster'] = kmeans.fit_predict(X_scaled)
penguins.sample(n=5, random_state=42)