PDF Improving_Promo...pdf | Request a copy Full text restricted Available under University of Tasmania Standard License.

Abstract

Promoter prediction is a well known, but challenging problem in the field of computational biology. Eukaryotic promoter prediction, an important step in the elucidation of transcriptional control networks and gene finding, is frustrated by the complex nature of promoters themselves. Within this paper we explore a representational scheme that describes promoters based on a variable number of salient binding sites within them. The multiple instance learning paradigm is used to allow these variable length instances to be reasoned about in a supervised learning context. We demonstrate that the procedure performs reasonably on its own, and allows for a significant increase in predictive accuracy when combined with physico-chemical promoter prediction.

Item Type:	Book Section
Authors/Creators:	Uren, PJ and Cameron-Jones, RM and Sale, AHJ
Keywords:	AI HCI Web intelligence agent technology artificial intelligence association rules case-based reasoning classification clustering cognitive technologies computational intelligence computer vision constraint satisfaction data mining decision making evolutionary computing fuzzy sets genetic algorithms image processing information extraction intelligent information systems internet security knowledge representation machine learning modal logic motion analysis multi-agent systems multi-instance learning natural language processing neural networks ontology optimization pattern recognition probabilistic methods question answering reinforcement learning robotics segmentation semantic Web
Journal or Publication Title:	Lecture Notes in Artificial Intelligence
Publisher:	Springer Verlag
ISSN:	978-3-540-89377-6
DOI / ID Number:	10.1007/978-3-540-89378-3_28
Related URLs:	Publisher
Item Statistics:	View statistics for this item

Actions (login required)

Item Control Page