计算机应用与软件2017，Vol.34Issue(6)：22-26,119,6.DOI:10.3969/j.issn.1000-386x.2017.06.005
基于稀疏正则化的高维数据可视化分析技术
HIGH-DIMENSIONAL DATA VISUALIZATION ANALYSIS TECHNOLOGY BASED ON SPARSE REGULARIZATION
摘要
Abstract
High-dimensional data visualization analysis is the research hotspot in the field of data analysis and visualization, the traditional low-dimensional dimension reduction method is often difficult to explain, and is not conducive to the visualization of high-dimensional data analysis and exploration.In this paper, a new visual explorer (Explainer) method is proposed to introduce the L1 sparse regularization feature selection into the high-dimensional data visualization process, and establish the relationship between high-level semantic tags and a few key features.The feasibility of the method is verified by visual design and experiment.It can improve the visualization performance of high dimensional data effectively.关键词
高维数据/特征选取/稀疏学习/可视化分析/降维/投影Key words
high-dimension data/Feature selection/Sparse learning/Visualization analysis/Dimension reduction/Projection分类
信息技术与安全科学引用本文复制引用
陈海辉,周向东,施伯乐..基于稀疏正则化的高维数据可视化分析技术[J].计算机应用与软件,2017,34(6):22-26,119,6.基金项目
国家自然科学基金项目(61370157) （61370157）
上海市科技项目(14511107403) （14511107403）
国网科技项目(5209401600 0A). （5209401600 0A）
