hartigan-kmeans

Hartigan's K-Means

Scope

This project provides an efficient implementation of Hartigan’s method for k-means clustering (Hartigan 1975). It builds on the work of Slonim, Aharoni and Crammer (2013), which introduced a significant improvement to the algorithm computational complexity, and adds an additional optimization for inputs in sparse vector representation. The project is packaged as a python library with a cython-wrapped C++ extension for the partition optimization code. A pure python implementation is included as well.

Installation

pip install hartigan-kmeans

Usage

The main class in this library is HKmeans, which implements the clustering interface of SciKit Learn, providing methods such as fit(), fit_transform(), fit_predict(), etc.

The sample code below clusters the 18.8K documents of the 20-News-Groups dataset into 20 clusters:

import numpy as np
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.datasets import fetch_20newsgroups
from sklearn import metrics
from hkmeans import HKMeans

# read the dataset
dataset = fetch_20newsgroups(subset='all', categories=None,
                             shuffle=True, random_state=256)

gold_labels = dataset.target
n_clusters = np.unique(gold_labels).shape[0]

# create count vectors using the 10K most frequent words
vectorizer = TfidfVectorizer(max_features=10000)
X = vectorizer.fit_transform(dataset.data)

# HKMeans initialization and clustering; parameters:
# perform 10 random initializations (n_init=10); the best one is returned.
# up to 15 optimization iterations in each initialization (max_iter=15)
# use all cores in the running machine for parallel execution (n_jobs=-1)
hkmeans = HKMeans(n_clusters=n_clusters, random_state=128, n_init=10,
                  n_jobs=-1, max_iter=15, verbose=True)
hkmeans.fit(X)

# report standard clustering metrics
print("Homogeneity: %0.3f" % metrics.homogeneity_score(gold_labels, hkmeans.labels_))
print("Completeness: %0.3f" % metrics.completeness_score(gold_labels, hkmeans.labels_))
print("V-measure: %0.3f" % metrics.v_measure_score(gold_labels, hkmeans.labels_))
print("Adjusted Rand-Index: %.3f" % metrics.adjusted_rand_score(gold_labels, hkmeans.labels_))
print("Adjusted Mutual-Info: %.3f" % metrics.adjusted_mutual_info_score(gold_labels, hkmeans.labels_))

Expected result:

Homogeneity: 0.245
Completeness: 0.290
V-measure: 0.266
Adjusted Rand-Index: 0.099
Adjusted Mutual-Info: 0.263

See the Examples directory for more illustrations and a comparison against Lloyd's K-Means.

License

Copyright IBM Corporation 2021

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

If you would like to see the detailed LICENSE click here.

Authors

• Algorithm: Hartigan 1975
• Pseudo-code and optimization: Slonim, Aharoni and Crammer (2013)
• Programming, optimization and maintenance: Assaf Toledo

If you have any questions or issues you can create a new issue here.

References

• Hartigan, John A. Clustering algorithms. Wiley series in probability and mathematical statistics: Applied probability and statistics. John Wiley & Sons, Inc., 1975.
• Slonim, Noam, Ehud Aharoni, and Koby Crammer. "Hartigan's K-Means Versus Lloyd's K-Means—Is It Time for a Change?." Twenty-Third International Joint Conference on Artificial Intelligence. 2013.