Hashing-based clustering in high dimensional data
详细信息 查看全文
- 作者：Juan Zamoraa ; juan.zamora@usm.cl" class="auth_mail" title="E-mail the corresponding author ; Marcelo Mendoza ; b ; mmendoza@inf.utfsm.cl" class="auth_mail" title="E-mail the corresponding author ; marcelo.mendoza@usm.cl" class="auth_mail" title="E-mail the corresponding author ; Hé ; ctor Allendea ; hallende@inf.utfsm.cl" class="auth_mail" title="E-mail the corresponding author
- 关键词：Locality sensitive hashing ; High dimensional clustering ; Min-wise hashing ; Random hyperplanes
- 刊名：Expert Systems with Applications
- 出版年：2016
- 出版时间：15 November 2016
- 年：2016
- 卷：62
- 期：Complete
- 页码：202-211
- 全文大小：518 K
文摘
- •
We modify hashing strategies to cluster high dimensional documents.
- •
We estimate the Jaccard similarity by counting bucket collisions between documents.
- •
We introduce a penalized Hamming function to approximate the cosine similarity.
- •
Both strategies allow improving the quality of the detected clusters.