Almost Linear Biobasis Function Neural Networks
2007 (English)In: The 2007 International Joint Conference on Neural Networks: IJCNN 2007 conference proceedings : August 12-17, 2007, Resaissance Orlando Resort, Orlando, Florida, USA, Piscataway, N.J.: IEEE Press, 2007, p. 1774-1778Conference paper, Published paper (Other academic)
Abstract [en]
An analysis of biobasis function neural networks is presented, which shows that the similarity metric used is a linear function and that bio-basis function neural networks therefore often end up being just linear classifiers in high dimensional spaces. This is a consequence of four things: the linearity of the distance measure, the normalization of the distance measure, the recommended default values of the parameters, and that biological data sets are sparse.
Place, publisher, year, edition, pages
Piscataway, N.J.: IEEE Press, 2007. p. 1774-1778
Series
International Conference on Neural Networks, ISSN 1098-7576 ; 2007
Keywords [en]
biobasis function neural networks, biological data set, biology computing, data analysis, distance measure linearity, distance measure normalization, linear classifiers, linear function, neural nets, pattern classification, similarity metric
National Category
Engineering and Technology
Identifiers
URN: urn:nbn:se:hh:diva-5021DOI: 10.1109/IJCNN.2007.4371226ISI: 000254291101124Scopus ID: 2-s2.0-51749115099ISBN: 978-1-4244-1380-5 OAI: oai:DiVA.org:hh-5021DiVA, id: diva2:327085
Conference
The 2007 International Joint Conference on Neural Networks
Note
©2007 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
2010-06-282010-06-282018-03-23Bibliographically approved