Batch Steepest-Descent-Mildest-Ascent for Interactive Maximum Margin Clustering

The maximum margin clustering principle extends support vector machines to unsupervised scenarios. We present a variant of this clustering scheme that can be used in the context of interactive clustering scenarios. In particular, our approach permits the class ratios to be manually defined by the us...

Full description

Saved in:
Bibliographic Details
Published inAdvances in Intelligent Data Analysis XIV Vol. 9385; pp. 95 - 107
Main Authors Gieseke, Fabian, Pahikkala, Tapio, Heskes, Tom
Format Book Chapter
LanguageEnglish
Published Switzerland Springer International Publishing AG 2015
Springer International Publishing
SeriesLecture Notes in Computer Science
Subjects
Artificial intelligence
Balance Constraint
Cluster Partition
Cluster Solution
Image processing
Interactive Cluster
Pattern recognition
Support Vector Machine
Online AccessGet full text
ISBN3319244647
9783319244648
ISSN0302-9743
1611-3349
DOI10.1007/978-3-319-24465-5_9

Cover

More Information
Summary:The maximum margin clustering principle extends support vector machines to unsupervised scenarios. We present a variant of this clustering scheme that can be used in the context of interactive clustering scenarios. In particular, our approach permits the class ratios to be manually defined by the user during the fitting process. Our framework can be used at early stages of the data mining process when no or very little information is given about the true clusters and class ratios. One of the key contributions is an adapted steepest-descent-mildest-ascent optimization scheme that can be used to fine-tune maximum margin clustering solutions in an interactive manner. We demonstrate the applicability of our approach in the context of remote sensing and astronomy with training sets consisting of hundreds of thousands of patterns.
ISBN:3319244647
9783319244648
ISSN:0302-9743
1611-3349
DOI:10.1007/978-3-319-24465-5_9