Sparse computation for large-scale binary classification

Baumann, Philipp (2014). Sparse computation for large-scale binary classification (Unpublished). In: IFORS 2014 : 20th Conference of the International Federation of Operational Research Societies. Barcelona. 13.-18.07.2014.

Well-known data mining algorithms rely on inputs in the form of pairwise similarities between objects. For large datasets it is computationally impossible to perform all pairwise comparisons. We therefore propose a novel approach that uses approximate Principal Component Analysis to efficiently identify groups of similar objects. The effectiveness of the approach is demonstrated in the context of binary classification using the supervised normalized cut as a classifier. For large datasets from the UCI repository, the approach significantly improves run times with minimal loss in accuracy.

Item Type:

Conference or Workshop Item (Speech)

Division/Institute:

03 Faculty of Business, Economics and Social Sciences > Department of Business Management > Institute of Financial Management > Professorship for Quantitative Methods in Business Administration

UniBE Contributor:

Baumann, Philipp

Subjects:

600 Technology > 650 Management & public relations

Language:

English

Submitter:

Larissa Notz

Date Deposited:

19 Nov 2014 09:52

Last Modified:

31 Jul 2017 08:19

URI:

https://boris.unibe.ch/id/eprint/59726

Actions (login required)

Edit item Edit item
Provide Feedback