Modelling relational statistics with Bayes Nets

Schulte, Oliver, Khosravi, Hassan, Kirkpatrick, Arthur E., Gao, Tianxiang and Zhu, Yuke (2014) Modelling relational statistics with Bayes Nets. Machine Learning, 94 1: 105-125. doi:10.1007/s10994-013-5362-7

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads

Author Schulte, Oliver
Khosravi, Hassan
Kirkpatrick, Arthur E.
Gao, Tianxiang
Zhu, Yuke
Title Modelling relational statistics with Bayes Nets
Journal name Machine Learning   Check publisher's open access policy
ISSN 0885-6125
Publication date 2014-01-01
Year available 2014
Sub-type Article (original research)
DOI 10.1007/s10994-013-5362-7
Open Access Status Not yet assessed
Volume 94
Issue 1
Start page 105
End page 125
Total pages 21
Place of publication New York, United States
Publisher Springer
Language eng
Abstract Class-level models capture relational statistics over object attributes and their connecting links, answering questions such as "what is the percentage of friendship pairs where both friends are women?" Class-level relationships are important in themselves, and they support applications like policy making, strategic planning, and query optimization. We represent class statistics using Parametrized Bayes Nets (PBNs), a first-order logic extension of Bayes nets. Queries about classes require a new semantics for PBNs, as the standard grounding semantics is only appropriate for answering queries about specific ground facts. We propose a novel random selection semantics for PBNs, which does not make reference to a ground model, and supports class-level queries. The parameters for this semantics can be learned using the recent pseudo-likelihood measure (Schulte in SIAM SDM, pp. 462-473, 2011) as the objective function. This objective function is maximized by taking the empirical frequencies in the relational data as the parameter settings. We render the computation of these empirical frequencies tractable in the presence of negated relations by the inverse Möbius transform. Evaluation of our method on four benchmark datasets shows that maximum pseudo-likelihood provides fast and accurate estimates at different sample sizes.
Keyword Bayes Nets
Möbius transform
Statistical-relational learning
Structured data
Q-Index Code C1
Q-Index Status Provisional Code
Institutional Status Non-UQ

Document type: Journal Article
Sub-type: Article (original research)
Collection: Institute for Teaching and Learning Innovation Publications
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 2 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 6 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Thu, 15 Sep 2016, 02:22:20 EST by Hassan Khosravi on behalf of Learning and Research Services (UQ Library)