Skip to content
Project: Clustering Antarctic Penguin Species
  • AI Chat
  • Code
  • Report
  • Arctic Penguin Exploration: Unraveling Clusters in the Icy Domain with K-means clustering

    Alt text 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!

    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.

    • 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 three species that are native to the region: Adelie, Chinstrap, and Gentoo, so 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.decomposition import PCA
    from sklearn.cluster import KMeans
    from sklearn.preprocessing import StandardScaler
    
    # Loading and examining the dataset
    penguins_df = pd.read_csv("data/penguins.csv")
    penguins_df.info()
    penguins_df['sex'].unique()
    penguins_df[(penguins_df['sex']=='.') | (penguins_df['sex'].isna())]
    penguins_df.boxplot()
    plt.xticks(rotation=90)
    plt.show()
    
    penguins_clean=penguins_df.dropna()
    penguins_clean[(penguins_clean['flipper_length_mm']>4000) | (penguins_clean['flipper_length_mm']<0)]
    penguins_clean=penguins_clean.drop([9,14])
    
    #Pre-processing using standard scaling, one-hot encoding
    
    ## create dummy variables
    
    df = pd.get_dummies(penguins_clean).drop('sex_.',axis=1)
    scaler = StandardScaler()
    X = scaler.fit_transform(df)
    penguins_preprocessed = pd.DataFrame(data=X,columns=df.columns)
    penguins_preprocessed.head()
    
    #PCA time
    pca = PCA(n_components=None)
    dfx_pca = pca.fit(penguins_preprocessed)
    n_components=sum(dfx_pca.explained_variance_ratio_>0.1)
    
    pca = PCA(n_components=n_components)
    penguins_PCA = pca.fit_transform(penguins_preprocessed)
    
    # Find optimal number of clusters using KMeans clustering
    
    inertia = []
    for k in range(1,10):
        kmeans = KMeans(n_clusters=k,random_state=42).fit(penguins_PCA)
        inertia.append(kmeans.inertia_)
    plt.plot(range(1,10),inertia,marker='o')
    plt.show()
    
    n_clusters=4
    
    #run the k means clustering with 4 clusters and visualise them
    KMeans(n_clusters=n_clusters,random_state=42).fit(penguins_PCA)
    plt.scatter(penguins_PCA[:,0],penguins_PCA[:,1],c=kmeans.labels_,cmap='viridis')
    plt.xlabel('First Principal Component')
    plt.ylabel('Second Principal Component')
    plt.title(f'K-means clustering(K={n_clusters})')
    plt.legend()
    plt.show()
    
    penguins_clean['label']= kmeans.labels_
    numeric_columns=['culmen_length_mm','culmen_depth_mm','flipper_length_mm','label']
    stat_penguins=penguins_clean[numeric_columns].groupby('label').mean()
    stat_penguins
    
    
    # Import Required Packages
    import pandas as pd
    import matplotlib.pyplot as plt
    from sklearn.decomposition import PCA
    from sklearn.cluster import KMeans
    from sklearn.preprocessing import StandardScaler
    
    # Step 1 - Loading and examining the dataset
    penguins_df = pd.read_csv("data/penguins.csv")
    penguins_df.head()
    penguins_df.info()
    
    # Step 2 - Dealing with null values and outliers
    penguins_df.boxplot()  
    plt.show()
    
    penguins_clean = penguins_df.dropna()
    penguins_clean[penguins_clean['flipper_length_mm']>4000]
    penguins_clean[penguins_clean['flipper_length_mm']<0]
    penguins_clean = penguins_clean.drop([9,14])
    
    # Step 3 - Perform preprocessing steps on the dataset to create dummy variables
    df = pd.get_dummies(penguins_clean).drop('sex_.',axis=1)
    
    # Step 4 - Perform preprocessing steps on the dataset - scaling
    scaler = StandardScaler()
    X = scaler.fit_transform(df)
    penguins_preprocessed = pd.DataFrame(data=X,columns=df.columns)
    penguins_preprocessed.head(10)
    
    # Step 5 - Perform PCA
    pca = PCA(n_components=None)
    dfx_pca = pca.fit(penguins_preprocessed)
    dfx_pca.explained_variance_ratio_
    n_components=sum(dfx_pca.explained_variance_ratio_>0.1)
    pca = PCA(n_components=n_components)
    penguins_PCA = pca.fit_transform(penguins_preprocessed)
    
    # Step 6 - Detect the optimal number of clusters for k-means clustering
    inertia = []
    for k in range(1, 10):
        kmeans = KMeans(n_clusters=k, random_state=42).fit(penguins_PCA)
        inertia.append(kmeans.inertia_)    
    plt.plot(range(1, 10), inertia, marker='o')
    plt.xlabel('Number of clusters')
    plt.ylabel('Inertia')
    plt.title('Elbow Method')
    plt.show()
    n_clusters=4
    
    # Step 7 - Run the k-means clustering algorithm
    # with the optimal number of clusters 
    # and visualize the resulting clusters.
    kmeans = KMeans(n_clusters=n_clusters, random_state=42).fit(penguins_PCA)
    plt.scatter(penguins_PCA[:, 0], penguins_PCA[:, 1], c=kmeans.labels_, cmap='viridis')
    plt.xlabel('First Principal Component')
    plt.ylabel('Second Principal Component')
    plt.title(f'K-means Clustering (K={n_clusters})')
    plt.legend()
    plt.show()
    
    # Step 8 - Create a final statistical DataFrame for each cluster.
    penguins_clean['label'] = kmeans.labels_
    numeric_columns = ['culmen_length_mm', 'culmen_depth_mm', 'flipper_length_mm','label']
    stat_penguins = penguins_clean[numeric_columns].groupby('label').mean()
    stat_penguins