These datasets have been generated from 4 LPADs:
shop.cpl is the shop example of LPAD of Meert, W., Struyf, J., Blockeel, H.: Learning ground cp-logic theories by leveraging
bayesian network learning techniques. Fundam. Inform. 89(1) (2008) 131–160. The other 3 are obtained by extending shop.
For each LPAD, the set of all its interpretation with non zero probability is computed and is used as training set, Each interpretation is annotated with its probability in the training set in order to represent its relative frequency.
The attached shop.zip file contains the 4 LPADs, their respective training sets plus the language bias files.
Moreover, for shop and shop3, a training set was generated by random sampling 1000 and 10,000 interpretations respectively. The file stem for these experiments is shop_sample and shop3_sample.