Clustering methods to detect fraud

Fraud Detection in Python

Charlotte Werger

Data Scientist

Clustering: trying to detect patterns in data

K-means clustering: using the distance to cluster centroids

K-means clustering in Python

# Import the packages
from sklearn.preprocessing import MinMaxScaler
from sklearn.cluster import KMeans

# Transform and scale your data
X = np.array(df).astype(np.float)

scaler = MinMaxScaler()
X_scaled = scaler.fit_transform(X)

# Define the k-means model and fit to the data
kmeans = KMeans(n_clusters=6, random_state=42).fit(X_scaled)

The right amount of clusters

Checking the number of clusters:

Silhouette method
Elbow curve

clust = range(1, 10) 
kmeans = [KMeans(n_clusters=i) for i in clust]

score = [kmeans[i].fit(X_scaled).score(X_scaled) for i in range(len(kmeans))]

plt.plot(clust,score)
plt.xlabel('Number of Clusters')
plt.ylabel('Score')
plt.title('Elbow Curve')
plt.show()

The elbow curve

Let's practice!

Fraud Detection in Python