Relational Information Gain

Marco Lippi, Manfred Jaeger, Paolo Frasconi, Andrea Passerini

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

5 Citationer (Scopus)


We introduce relational information gain, a refinement scoring function measuring the informativeness of newly introduced variables. The gain can be interpreted as a conditional entropy in a well-defined sense and can be efficiently approximately computed. In conjunction with simple greedy general-to-specific search algorithms such as FOIL, it yields an efficient and competitive algorithm in terms of predictive accuracy and compactness of the learned theory. In conjunction with the decision tree learner TILDE, it offers a beneficial alternative to lookahead, achieving similar performance while significantly reducing the number of evaluated literals
TidsskriftMachine Learning
Udgave nummer2
Sider (fra-til)219-239
Antal sider21
StatusUdgivet - 2011