论文题目：Robust Unsupervised Feature Selection viaMulti-Group Adaptive Graph Representation
作者：M. You#, A. Yuan#*, M. Zou, D. He and X. Li
期刊名称：IEEE Transactions on Knowledge and Data Engineering（CCFA，机器学习、数据挖掘领域top期刊）
发表时间：2021年11月
论文摘要：
Unsupervised feature selection can play an important role in addressing the issue of processing massive unlabeledhigh-dimensional data in the domain of machine learning and data mining. This paper presents a novel unsupervised feature selectionmethod, referred to as Multi-Group Adaptive Graph Representation (MGAGR). Different from existing methods, the relationshipbetween features is explored via the global similarity matrix, which is reconstructed by local similarities of multiple groups. Specifically,the similarity of a feature compared to other features can be represented by the linear combination of all the local similarities. The localsimilarity of a representative group is given a large weight to reconstruct the global similarity. Besides, an iterative algorithm is given tosolve the optimization problem, in which the global similarity matrix, its corresponding reconstruction weights and theself-representation matrix are iteratively improved. Experimental results on 8 benchmark datasets demonstrates that the proposedmethod outperforms the state-of-the-art unsupervised feature selection methods in terms of clustering performance. The source codeis available at: https://github.com/misteru/MGAGR.
论文链接：https://ieeexplore.ieee.org/document/9606609