We are interested in learning complex combinatorial features from relational data. We rely on an expressive and general representation language whose semantics allows us to express many features that have been used in different statistical relational learning settings. To avoid expensive exhaustive search over the space of relational features, we introduce a heuristic search algorithm guided by a generalized relational notion of information gain and a discriminant function. The algorithm succesfully finds interesting and interpretable features on artificial and real-world relational learning problems

Feature Discovery with Type Extension Trees / P. Frasconi ; M. Jaeger ; A. Passerini. - STAMPA. - 5194(2008), pp. 122-139. ((Intervento presentato al convegno 18th International Conference on Inductive Logic Programming, ILP 2008 [10.1007/978-3-540-85928-4_13].

Feature Discovery with Type Extension Trees

FRASCONI, PAOLO;
2008

Abstract

We are interested in learning complex combinatorial features from relational data. We rely on an expressive and general representation language whose semantics allows us to express many features that have been used in different statistical relational learning settings. To avoid expensive exhaustive search over the space of relational features, we introduce a heuristic search algorithm guided by a generalized relational notion of information gain and a discriminant function. The algorithm succesfully finds interesting and interpretable features on artificial and real-world relational learning problems
ILP 2008: proceedings
18th International Conference on Inductive Logic Programming, ILP 2008
Goal 4: Quality education
Goal 9: Industry, Innovation, and Infrastructure
P. Frasconi ; M. Jaeger ; A. Passerini
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/2158/387159
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 5
  • ???jsp.display-item.citation.isi??? 3
social impact