pipe = Pipeline([('scaler', PowerTransformer())])df_aux = pd.DataFrame(pipe_fit.fit_transform(df_no_outliers[["age", "balance"]] ), columns = ["age", "balance"])df_no_outliers_norm = df_no_outliers.copy()# Replace age and balance columns by preprocessed valuesdf_no_outliers_norm = df_no_outliers_norm.drop(["age", "balance"], axis = 1)df_no_outliers_norm["age"] = df_aux["age"].valuesdf_no_outliers_norm["balance"] = df_aux["balance"].valuesdf_no_outliers_norm

# Choose optimal K using Elbow methodfrom kmodes.kprototypes import KPrototypesfrom plotnine import *import plotninecost = []range_ =range(2, 15)for cluster in range_:kprototype = KPrototypes(n_jobs = -1, n_clusters = cluster, init = 'Huang', random_state = 0)kprototype.fit_predict(df_no_outliers, categorical = categorical_columns_index)cost.append(kprototype.cost_)print('Cluster initiation: {}'.format(cluster))# Converting the results into a dataframe and plotting themdf_cost = pd.DataFrame({'Cluster':range_, 'Cost':cost})# Data vizplotnine.options.figure_size = (8, 4.8)(ggplot(data = df_cost)+geom_line(aes(x ='Cluster',y ='Cost'))+geom_point(aes(x ='Cluster',y ='Cost'))+geom_label(aes(x ='Cluster',y ='Cost',label ='Cluster'),size =10,nudge_y =1000) +labs(title ='Optimal number of cluster with Elbow Method')+xlab('Number of Clusters k')+ylab('Cost')+theme_minimal())

# We get the index of categorical columnsnumerics = ['int16', 'int32', 'int64', 'float16', 'float32', 'float64']categorical_columns = df_no_outliers_norm.select_dtypes(exclude=numerics).columnsprint(categorical_columns)categorical_columns_index = [df_no_outliers_norm.columns.get_loc(col)for col in categorical_columns]# Create the modelcluster_num =5kprototype = KPrototypes(n_jobs = -1, n_clusters = cluster_num, init = 'Huang', random_state = 0)kprototype.fit(df_no_outliers_norm, categorical = categorical_columns_index)clusters = kprototype.predict(df_no_outliers , categorical = categorical_columns_index)print(clusters) " -> array([3, 1, 1, ..., 1, 1, 2], dtype=uint16)"

from prince import MCAdef get_MCA_3d(df, predict):mca = MCA(n_components =3, n_iter = 100, random_state = 101)mca_3d_df = mca.fit_transform(df)mca_3d_df.columns = ["comp1", "comp2", "comp3"]mca_3d_df["cluster"] = predictreturn mca, mca_3d_dfdef get_MCA_2d(df, predict):mca = MCA(n_components =2, n_iter = 100, random_state = 101)mca_2d_df = mca.fit_transform(df)mca_2d_df.columns = ["comp1", "comp2"]mca_2d_df["cluster"] = predictreturn mca, mca_2d_df"-------------------------------------------------------------------"mca_3d, mca_3d_df = get_MCA_3d(df_no_outliers_norm, clusters)

import pandas as pd # dataframe manipulationimport numpy as np # linear algebrafrom sentence_transformers import SentenceTransformerdf = pd.read_csv("data/train.csv", sep = ";")# -------------------- First Step --------------------def compile_text(x):text =f"""Age: {x['age']}, housing load:{x['housing']}, Job:{x['job']}, Marital:{x['marital']}, Education:{x['education']}, Default:{x['default']}, Balance:{x['balance']}, Personal loan:{x['loan']}, contact:{x['contact']}"""return textsentences = df.apply(lambda x: compile_text(x), axis=1).tolist()# -------------------- Second Step --------------------model = SentenceTransformer(r"sentence-transformers/paraphrase-MiniLM-L6-v2")output = model.encode(sentences=sentences,show_progress_bar=True,normalize_embeddings=True)df_embedding = pd.DataFrame(output)df_embedding

# Normal Datasetdf = pd.read_csv("data/train.csv", sep = ";")df = df.iloc[:,0:8]# Embedding Datasetdf_embedding = pd.read_csv("data/embedding_train.csv", sep = ",")

df_embedding_no_out.shape -> (40690, 384)df_embedding_with_out.shape -> (45211, 384)

n_clusters = 5clusters = KMeans(n_clusters=n_clusters, init ="k-means++").fit(df_embedding_no_out)print(clusters.inertia_)clusters_predict = clusters.predict(df_embedding_no_out)

Davies bouldin score: 1.8095386826791042Calinski Score: 6419.447089002081Silhouette Score: 0.20360442824114108

plot_pca_3d(df_pca_3d, title ="PCA Space", opacity=0.2, width_line = 0.1)