Supervised learning in the gene ontology part II: a bottom-up algorithm

Authors:
Herman Midelfart
Affiliations:
Department of Biology, Norwegian University of Science and Technology, Trondheim, Norway
Venue:
Transactions on Rough Sets IV
Year:
2005

Citing 7
Cited 3

Information storage and retrieval

Information storage and retrieval
Separate-and-Conquer Rule Learning

Artificial Intelligence Review
Modern Information Retrieval

Modern Information Retrieval
Genes, Themes, and Microarrays: Using Information Retrieval for Large-Scale Gene Analysis

Proceedings of the Eighth International Conference on Intelligent Systems for Molecular Biology
Classification of Gene Expression Data in an Ontology

ISMDA '01 Proceedings of the Second International Symposium on Medical Data Analysis
A Rough Set Framework for Learning in a Directed Acyclic Graph

TSCTC '02 Proceedings of the Third International Conference on Rough Sets and Current Trends in Computing
Supervised learning in the gene ontology part i: a rough set framework

Transactions on Rough Sets IV

Induction and selection of the most interesting Gene Ontology based multiattribute rules for descriptions of gene groups

Pattern Recognition Letters
Quality improvement of rule-based gene group descriptions using information about GO terms importance occurring in premises of determined rules

International Journal of Applied Mathematics and Computer Science
Supervised learning in the gene ontology part i: a rough set framework

Transactions on Rough Sets IV

Quantified Score

Hi-index	0.00

Visualization

Abstract

Prediction of gene function for expression profiles introduces a new problem for supervised learning algorithms. The decision classes are taken from an ontology, which defines relationships between the classes. Supervised algorithms, on the other hand, assumes that the classes are unrelated. Hence, we introduce a new algorithm which can take these relationships into account. This is tested on a microarray data set created from human fibroblast cells and on several artificial data sets. Since standard performance measures do not apply to this problem, we also introduce several new measures for measuring classification performance in an ontology.