文献学习(part5)--Clustering by Passing Messages Between Data Points

最新推荐文章于 2023-08-07 00:20:23 发布

GUI Research Group

最新推荐文章于 2023-08-07 00:20:23 发布

阅读量284

点赞数

分类专栏：学习日志机器学习文章标签：算法聚类

本文链接：https://blog.csdn.net/m0_37422217/article/details/109356798

版权

机器学习同时被 2 个专栏收录

413 篇文章 55 订阅 ¥99.90 ¥299.90

订阅专栏

学习日志

26 篇文章 18 订阅 ¥99.90 ¥299.90

订阅专栏

超级会员免费看

本文介绍了亲和传播算法，这是一种通过在数据点之间传递消息来进行聚类的方法，尤其适用于处理相似性数据。与传统的k-centers聚类相比，亲和传播在寻找高质量样本和簇方面表现更优，且计算时间更短。它不需要预先指定聚类数量，而是通过输入数据点之间的相似性来确定。此外，通过调整输入偏好，可以自动选择合适的模型。实验表明，亲和传播在人脸图像聚类和基因检测等任务中能获得更低的误差。

摘要由CSDN通过智能技术生成

学习笔记，仅供参考

文章目录

- Clustering by Passing Messages Between Data Points

Clustering by Passing Messages Between Data Points

通过在数据点之间传递消息进行聚类

Clustering data by identifying a subset of representative examples is important for processing sensory signals and detecting patterns in data. Such “exemplars” can be found by randomly choosing an initial subset of data points and then iteratively refining it, but this works well only if that initial choice is close to a good solution. We devised a method called “affinity propagation,” which takes as input measures of similarity between pairs of data points. Real-valued messages are exchanged between data points until a high-quality set of exemplars and correspo

了解本专栏