tfidf_matrix = tfidf_vectorizer.fit_transform(data)

from sklearn.metrics.pairwise import cosine_similarity
dist = 1 - cosine_similarity(tfidf_matrix)

from scipy.cluster.hierarchy import ward, dendrogram, fcluster

linkage_matrix = ward(dist)

#Lowest level clustering
clusters = fcluster(Z = linkage_matrix,t = 10, criterion='maxclust')