What are SVMs?

From WormBaseWiki
Revision as of 13:08, 20 November 2009 by Vanaukenk (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search

SVMs (Support Vector Machines) are machine learning algorithms that can be used to classify objects.

In the context of first pass curation, SVMs are used to classify documents with respect to the type of data they contain, for example RNAi experiments, sequence of mutant alleles, or expression patterns.

SVM classification systems use features, i.e., characteristics of the objects to be classified, in our case words, numbers or phrases in a paper, to identify characteristic features of a particular data type and then use those features to make predictions about new object, in our case new papers.

As machine learning algorithms, SVMs require positive and negative training sets to effectively train the algorithm.


SVM Guidelines

Caltech documentation