Trace Ratio Criterion for Feature Selection

By Feiping Nie, Shiming Xiang, Yangqing Jia, Changshui Zhang, Shuicheng Yan. AAAI Conference on Artificial Intelligence, 2008

378 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/8bc814c9653ef7fe248986788dd2a53375317a3a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Jundong Li
2. Jiliang Tang
3. Huan Liu
4. Charu C. Aggarwal
5. Feiping Nie
6. Chenping Hou
7. Zheng Zhao
8. Marcus Chen

Citations in the corpus

• Reconstruction-based Unsupervised Feature Selection: An Embedded Approach (2017, methodology): “…take advantage of the discriminative information encoded in class labels to select the subset of features that are able to distinguish instances from different classes [Tibshirani, 1996; Ding and Peng, 2005; Nie et al., 2008; 2010; Cover and Thomas, 2012; Jian et al., 2016; Li et al., 2017].”
• Generalized Fisher Score for Feature Selection (2011, methodology): “We compare the proposed method to the state-of-art feature selection methods: Fisher Score [15], Laplacian Score [6], Hilbert Schmidt Independence Criterion (HSIC) [18] and Trace Ratio criterion [12].”
• Efficient Spectral Feature Selection with Minimum Redundancy (2010, methodology): “There exist many spectral feature selection algorithms: Laplacian score (He, Cai, and Niyogi 2005), Fisher score (Duda, Hart, and Stork 2001), trace ratio (Nie et al. 2008), relief and reliefF (Sikonja and Kononenko 2003), SPEC (Zhao and Liu 2007), and HSIC (Song et al. 2007).”
• A Multimedia Retrieval Framework Based on Semi-Supervised Ranking and Relevance Feedback (2012, unknown): “In Algorithm-3, we have leveraged the trick in [23] to accelerate the convergence speed.”
• Feature Selection (2016, methodology): “4 Trace Ratio Criterion (Nie et al., 2008) (Supervised) Recently, the trace ratio criterion has been proposed to directly select the global optimal feature subset based on the corresponding score, which is computed in a trace ratio norm.”
• Joint Embedding Learning and Sparse Regression: A Framework for Unsupervised Feature Selection (2014, methodology): “Most unsupervised feature selection approaches are either based on filters [13], [14], [15], wrappers [4], [16] or embedding [17], [18], [19].”
• On Similarity Preserving Feature Selection (2013, methodology): “In [28], with different definitions of L b and L w , the trace-ratio criterion generalizes Fisher Score and Lapalacian Score to evaluate a feature subset as a whole.”
• A Unified Feature Selection Framework for Graph Embedding on High Dimensional Data (2015, methodology): “[23] proposed a feature subset trace ratio method to rank feature subsets that best discriminate between-class samples and preserve withinclass relationships.”
• Pattern Recognition (2018, unknown): “These existing algorithms often use different frameworks for the two tasks.”
• A Convolutional Learning System for Object Classification in 3-D Lidar Data (2010, methodology): “D CNN with the ability to merge more features with progressively decreasing resolution and use more combinations of features for object classification.”