python – sklearn分解顶级术语

2023年3月13日 207次阅读

有没有办法在数据分解时确定每个群集的主要功能/术语？

在sklearn文档中的示例中,通过对特征进行排序并与具有相同数量的特征的矢量化器feature_names进行比较来提取顶部术语.

http://scikit-learn.org/stable/auto_examples/document_classification_20newsgroups.html

我想知道如何实现get_top_terms_per_cluster()：

X = vectorizer.fit_transform(dataset)  # with m features
X = lsa.fit_transform(X)  # reduce number of features to m'
k_means.fit(X)
get_top_terms_per_cluster()  # out of m features

最佳答案 对于某些k,假设lsa = TruncatedSVD(n_components = k),获得项权重的明显方法是利用LSA / SVD是线性变换的事实,即lsa.components_的每一行是输入项的加权和. ,你可以将它与k-means中的聚类质心相乘.

让我们设置一些东西并训练一些模型：

>>> from sklearn.datasets import fetch_20newsgroups
>>> from sklearn.feature_extraction.text import TfidfVectorizer
>>> from sklearn.cluster import KMeans
>>> from sklearn.decomposition import TruncatedSVD
>>> data = fetch_20newsgroups()
>>> vectorizer = TfidfVectorizer(min_df=3, max_df=.95, stop_words='english')
>>> lsa = TruncatedSVD(n_components=10)
>>> km = KMeans(n_clusters=3)
>>> X = vectorizer.fit_transform(data.data)
>>> X_lsa = lsa.fit_transform(X)
>>> km.fit(X_lsa)

现在乘以LSA分量和k均值质心：

>>> X.shape
(11314, 38865)
>>> lsa.components_.shape
(10, 38865)
>>> km.cluster_centers_.shape
(3, 10)
>>> weights = np.dot(km.cluster_centers_, lsa.components_)
>>> weights.shape
(3, 38865)

然后打印;由于LSA中的符号不​​确定性,我们需要权重的绝对值：

>>> features = vectorizer.get_feature_names()
>>> weights = np.abs(weights)
>>> for i in range(km.n_clusters):
...     top5 = np.argsort(weights[i])[-5:]
...     print(zip([features[j] for j in top5], weights[i, top5]))
...     
[(u'escrow', 0.042965734662740895), (u'chip', 0.07227072329320372), (u'encryption', 0.074855609122467345), (u'clipper', 0.075661844826553887), (u'key', 0.095064798549230306)]
[(u'posting', 0.012893125486957332), (u'article', 0.013105911161236845), (u'university', 0.0131617377000081), (u'com', 0.023016036009601809), (u'edu', 0.034532489348082958)]
[(u'don', 0.02087448155525683), (u'com', 0.024327099321009758), (u'people', 0.033365757270264217), (u'edu', 0.036318114826463417), (u'god', 0.042203130080860719)]

请注意,你真的需要一个停用词过滤器来实现这一点.停用词往往会在每个组件中结束,并在每个群集质心中获得高权重.

点赞