Deep Embedding for Determining the Number of Clusters
Abstract
Determining the number of clusters is important but challenging, especially for data of high dimension. In this paper, we propose Deep Embedding Determination (DED), a method that can solve jointly for the unknown number of clusters and feature extraction. DED first combines the virtues of the convolutional autoencoder and the t-SNE technique to extract low dimensional embedded features. Then it determines the number of clusters using an improved density-based clustering algorithm. Our experimental evaluation on image datasets shows significant improvement over state-of-the-art methods and robustness with respect to hyperparameter settings.
Cite
Text
Wang et al. "Deep Embedding for Determining the Number of Clusters." AAAI Conference on Artificial Intelligence, 2018. doi:10.1609/AAAI.V32I1.12150Markdown
[Wang et al. "Deep Embedding for Determining the Number of Clusters." AAAI Conference on Artificial Intelligence, 2018.](https://mlanthology.org/aaai/2018/wang2018aaai-deep-a/) doi:10.1609/AAAI.V32I1.12150BibTeX
@inproceedings{wang2018aaai-deep-a,
title = {{Deep Embedding for Determining the Number of Clusters}},
author = {Wang, Yiqi and Shi, Zhan and Guo, Xifeng and Liu, Xinwang and Zhu, En and Yin, Jianping},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2018},
pages = {8173-8174},
doi = {10.1609/AAAI.V32I1.12150},
url = {https://mlanthology.org/aaai/2018/wang2018aaai-deep-a/}
}