As FOIL, GOLEM is a “classic” among empirical ILP systems. It has been applied successfully on real-world problems such as protein structure prediction and finite element mesh design. GOLEM copes efficiently with large datasets. It achieves this efficiency because it avoids searching a large hypothesis space for consistent hypotheses as, for instance, FOIL, but rather constructs a unique clause covering a set of positive examples relative to the available background knowledge. The principle is based on the relative least general generalisations (rlggs) introduced by Plotkin. GOLEM embeds the construction of rlggs in a covering approach. For the induction of a single clause, it randomly selects several pairs of positive examples and computes their rlggs. Among these rlggs, GOLEM chooses the one which covers the largest number of positive examples and is consistent with the negative examples. This clause is then further generalised. GOLEM randomly selects a set of positive examples and constructs the rlggs of each of these examples and the clause obtained in the first construction step. Again, the rlgg with the greatest coverage is selected and generalised by the same process. The generalisation process is repeated until the coverage of the best clause stops increasing. GOLEM conducts a postprocessing step, which reduces induced clauses by removing irrelevant literals. ..

References in zbMATH (referenced in 53 articles )

Showing results 1 to 20 of 53.
Sorted by year (citations)

1 2 3 next

  1. Cropper, Andrew; Dumančić, Sebastijan; Evans, Richard; Muggleton, Stephen H.: Inductive logic programming at 30 (2022)
  2. Cropper, Andrew; Tourret, Sophie: Logical reduction of metarules (2020)
  3. Cropper, Andrew; Muggleton, Stephen H.: Learning efficient logic programs (2019)
  4. Muggleton, Stephen H.; Schmid, Ute; Zeller, Christina; Tamaddoni-Nezhad, Alireza; Besold, Tarek: Ultra-strong machine learning: comprehensibility of programs learned with ILP (2018)
  5. Angiulli, Fabrizio; Fassetti, Fabio: Exploiting domain knowledge to detect outliers (2014)
  6. Muggleton, Stephen; De Raedt, Luc; Poole, David; Bratko, Ivan; Flach, Peter: ILP turns 20. Biography and future challenges (2012)
  7. Srinivasan, Ashwin; Faruquie, Tanveer A.; Joshi, Sachindra: Data and task parallelism in ILP using mapreduce (2012)
  8. Sakama, Chiaki; Inoue, Katsumi: Inductive equivalence in clausal logic and nonmonotonic logic programming (2011)
  9. Tamaddoni-Nezhad, Alireza; Muggleton, Stephen: Stochastic refinement (2011)
  10. Muggleton, Stephen; Santos, José; Tamaddoni-Nezhad, Alireza: ProGolem: a system based on relative minimal generalisation (2010)
  11. Fonseca, Nuno A.; Srinivasan, Ashwin; Silva, Fernando; Camacho, Rui: Parallel ILP for distributed-memory architectures (2009)
  12. Kitzelmann, Emanuel: Analytical inductive functional programming (2009)
  13. Tamaddoni-Nezhad, Alireza; Muggleton, Stephen: The lattice structure and refinement operators for the hypothesis space bounded by a bottom clause (2009)
  14. Dietterich, Thomas G.; Domingos, Pedro; Getoor, Lise; Muggleton, Stephen; Tadepalli, Prasad: Structured machine learning: the next ten years (2008)
  15. Guo, Hongyu; Viktor, Herna L.: Multirelational classification: a multiple view approach (2008) ioport
  16. Muggleton, Stephen; Tamaddoni-Nezhad, Alireza: QG/GA: a stochastic search for Progol (2007)
  17. Arias, Marta; Khardon, Roni: Complexity parameters for first order classes (2006) ioport
  18. Arias, Marta; Khardon, Roni: Complexity parameters for first order classes (2006)
  19. Cicekli, Ilyas; Cicekli, Nihan Kesim: Generalizing predicates with string arguments (2006)
  20. Cicekli, Ilyas; Cicekli, Nihan Kesim: Generalizing predicates with string arguments (2006) ioport

1 2 3 next