首页 > 其他 > 详细

PP: Pattern Trails: visual analysis of pattern transitions in subspaces

时间:2020-01-15 09:06:33      阅读:84      评论:0      收藏:0      [点我收藏+]

Problem:

1. We can‘t find patterns in full attribute space, and patterns may only be found in smaller subspaces. 

2. Pattern Trails is an interactive visual approach for the exploration of subspaces of multivariate data

Related work

1. Multivariate data analysis and visualization

parallel coordinate plots; pixel bar charts; Chernoff faces.

2. using dimensionality reduction for visual analysis

 MDS: multidimensional scaling;

PCA: principal component analysis; 

t-SNE: t-distributed stochastic neighbor embedding;

SOM: self-organizing maps

3. Subspace search and visualization

Methodology

1. Derive interesting subspaces of the multivariate data

 

Supplementary knowledge

1. what are subspace patterns and subspace analysis. 

2. The main task in understanding such multivariate data is to identify and interpret relevant patterns like dense groups (clusters), outliers, or correlations.

3. Techniques for visually exploring multivariate data:

Parallel Coordinate Plots;

dimensionality reduction降维; to transform the data to a lower-dimensional space but preserving the main structure of the data. 

4. Chernoff Faces

5. 

 

PP: Pattern Trails: visual analysis of pattern transitions in subspaces

原文:https://www.cnblogs.com/dulun/p/12194732.html

(0)
(0)
   
举报
评论 一句话评论(0
关于我们 - 联系我们 - 留言反馈 - 联系我们:wmxa8@hotmail.com
© 2014 bubuko.com 版权所有
打开技术之扣,分享程序人生!