sklearn文本特征预处理2:Similarity, 聚类, LDA, word2vec

接上一篇<sklearn文本特征预处理1: WordPunctTokenizer, CountVectorizer, TF-IDF>

五. Similarity特征

# 余弦相似度
from sklearn.metrics.pairwise import cosine_similaritysimilarity_matrix = cosine_similarity(tv_matrix)
similarity_df = pd.DataFrame(similarity_matrix)
similarity_df

六. 聚类特征

from sklearn.cluster import KMeanskm = KMeans(n_clusters = 2)
km.fit_transform(similarity_df)
cluster_labels = km.labels_
cluster_labels = pd.DataFrame(cluster_labels, columns=['ClusterLabel'])
pd.concat