Visualization and interactive feature selection for unsupervised data

By Jennifer G. Dy, Carla E. Brodley. 2000

70 citations recorded by the source corpus.

For many feature selection problems, a human de nes the features that are potentially useful, and then a subset is chosen from the original pool of features using an automated feature selection algorithm. In contrast to supervised learning, class information is not available to guide the feature search for unsupervised learning tasks. In this paper, we introduce Visual-FSSEM (Visual Feature Subset Selection using Expectation-Maximization Clustering), which incorporates visualization techniques, clustering, and user interaction to guide the feature subset search and to enable a deeper understanding of the data. Visual-FSSEM, serves both as an exploratory and multivariate-data visualization tool. We illustrate Visual-FSSEM on a high-resolution computed tomography lung image data set.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning, Human-Computer Interaction, Signal Processing

How researchers see this paper

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

In Artificial Intelligence

1. Huan Liu
2. Jun-Yan Zhu
3. David Bau
4. Manoranjan Dash
5. Antonio Torralba
6. Ki-Hoon Choi
7. Peter Scheuermann
8. Tongzhou Wang

In Computer Vision and Pattern Recognition

1. David Bau
2. Jun-Yan Zhu
3. Antonio Torralba
4. Huan Liu
5. Luc Van Gool
6. Dacheng Tao
7. Yi Yang
8. Ming-Hsuan Yang

In Representation & Self-Supervised Learning

1. David Bau
2. Ki-Hoon Choi
3. Peter Scheuermann
4. Manoranjan Dash
5. Huan Liu
6. Jennifer G. Dy
7. Carla E. Brodley
8. Jun-Yan Zhu

In Human-Computer Interaction

1. Ki-Hoon Choi
2. Peter Scheuermann
3. Manoranjan Dash
4. Huan Liu
5. Jennifer G. Dy
6. Carla E. Brodley
7. Antonio Torralba
8. Ali Borji

In Signal Processing

1. David Bau
2. Jun-Yan Zhu
3. Ki-Hoon Choi
4. Peter Scheuermann
5. Manoranjan Dash
6. Huan Liu
7. Jennifer G. Dy
8. Carla E. Brodley

Citations in the corpus

• Rewriting a Deep Generative Model (2020, background): “Interactive machine learning systems aim to improve training through human interaction in labeling [15,22,68], or by allowing a user to to aid in the model optimization process via interactive feature selection [19,27,64,50] or model and hyperparameter selection [39,61,37].”
• From Visual Data Exploration to Visual Data Mining: A Survey (2003, background): “Dy and Brodley [25] tackle the feature selection problem, a step that usually precedes clustering to identify a feature subset that best discovers data clusters.”
• Feature Selection for Unsupervised Learning (2004, methodology): “In particular, this paper investigates the wrapper framework through FSSEM (feature subset selection using EM clustering) introduced in (Dy and Brodley, 2000a).”
• Feature selection for clustering - a filter solution (2002, methodology): “In [11] EM (Expectation–Maximization)and trace measure are used for evaluation.”