Uncertain groupings: probabilistic combination of grouping data

B. Wanders, M. van Keulen, P.E. van der Vet

Research output: Book/ReportReportProfessional

142 Downloads (Pure)

Abstract

A bioinformatician has a large number of homology data sources to choose from. These data sources need to be combined before a query can be posed over the combined data. We propose a generic probabilistic approach to combining grouping data from multiple sources. Our approach incorporates an iteratively evolving view on trust, allowing the bioinformatician to express his fine-grained view on how much the data in the sources can be trusted. We evaluate our approach by combining 3 real-world biological databases and show that it scales well for realistic amounts of data and uncertainty.
Original languageEnglish
Place of PublicationEnschede
PublisherCentre for Telematics and Information Technology (CTIT)
Number of pages10
Publication statusPublished - 5 Dec 2014

Publication series

NameCTIT Technical Report Series
PublisherUniversity of Twente, Centre for Telematics and Information Technology (CTIT)
No.TR-CTIT-14-12
ISSN (Print)1381-3625

Fingerprint

Dive into the research topics of 'Uncertain groupings: probabilistic combination of grouping data'. Together they form a unique fingerprint.

Cite this