Learning from structured data by finger printing
2006 (English)In: Publications of the Finnish Artificial Intelligence Society, 2006, 120-126 p.Conference paper (Refereed)
Current methods for learning from structured data are limited w.r.t. handling large or isolated substructures and also impose constraints on search depth and induced structure length. An approach to learning from structured data using a graph based canonical representation method of structures, called finger printing, is introduced that addresses the limitations of current methods. The method is implemented in a system, called DIFFER, which is demonstrated to compare favourable to existing state-of-art methods on some benchmark data sets. It is shown that further improvements can be obtained by combining the features generated by finger printing with features generated by previous methods.
Place, publisher, year, edition, pages
2006. 120-126 p.
, Publications of the Finnish Artificial Intelligence Society, ISSN 1796-623X
Benchmark data, Canonical representations, Finger printing, Graph-based, Induced structures, State-of-art methods, Structured data, Artificial intelligence, Printing
IdentifiersURN: urn:nbn:se:kth:diva-155040ScopusID: 2-s2.0-84862568162ISBN: 9525677001ISBN: 9789525677003OAI: oai:DiVA.org:kth-155040DiVA: diva2:759446
9th Scandinavian Conference on Artificial Intelligence, SCAI 2006, 25 October 2006 through 27 October 2006, Espoo
QC 201410302014-10-302014-10-292014-10-30Bibliographically approved