- author = "Tim Weninger and William H. Hsu and Jing Xia and Waleed Aljandal",
- title = "An evolutionary approach to constructive induction for link discovery",
- booktitle = "GECCO-2009 Late-Breaking Papers",
- year = "2009",
- editor = "Anna I. Esparcia and Ying-ping Chen and Gabriela Ochoa and Ender Ozcan and Marc Schoenauer and Anne Auger and Hans-Georg Beyer and Nikolaus Hansen and Steffen Finck and Raymond Ros and Darrell Whitley and Garnett Wilson and Simon Harding and W. B. Langdon and Man Leung Wong and Laurence D. Merkle and Frank W. Moore and Sevan G. Ficici and William Rand and Rick Riolo and Nawwaf Kharma and William R. Buckley and Julian Miller and Kenneth Stanley and Jaume Bacardit and Will Browne and Jan Drugowitsch and Nicola Beume and Mike Preuss and Stephen L. Smith and Stefano Cagnoni and Jim DeLeo and Alexandru Floares and Aaron Baughman and Steven Gustafson and Maarten Keijzer and Arthur Kordon and Clare Bates Congdon and Laurence D. Merkle and Frank W. Moore",
- pages = "2167--2172",
- address = "Montreal",
- publisher = "ACM",
- publisher_address = "New York, NY, USA",
- month = "8-12 " # jul,
- organisation = "SigEvo",
- keywords = "genetic algorithms, genetic programming",
- isbn13 = "978-1-60558-325-9",
- bibsource = "DBLP, http://dblp.uni-trier.de",
-
DOI = "
doi:10.1145/1570256.1570295",
- abstract = "This paper presents a genetic programming-based symbolic regression approach to the construction of relational features in link analysis applications. Specifically, we consider the problems of predicting, classifying and annotating friends relations in friends networks, based upon features constructed from network structure and user profile data. We first document a data model for the blog service LiveJournal, and define a set of machine learning problems such as predicting existing links and estimating inter-pair distance. Next, we explain how the problem of classifying a user pair in a social network, as directly connected or not, poses the problem of selecting and constructing relevant features. We use genetic programming to construct features, represented by multiple symbol trees with base features as their leaves. In this manner, the genetic program selects and constructs features that may not have been originally considered, but possess better predictive properties than the base features. Finally, we present classification results and compare these results with those of the control and similar approaches.",
-
notes = "PhD University of Illinois 22 Aug 2013
http://hdl.handle.net/2142/45402 Discovering roles and
types from hierarchical information networks (not
GP)
Distributed on CD-ROM at GECCO-2009.
ACM Order Number 910092.",
