Reconstructing True Wrong Inductions
Résumé
Many pre-scientific and common sense inductions are erroneous. Here are presented attempts to simulate that kind of inductive reasoning. Our hypothesis is that mistaken inductions are not only due to the lack of facts, but also to the poor description of existing facts and to implicit knowledge. We present a few experiments which aim at validating this hypothesis by simulating with machine learning and data mining techniques the way people erect erroneous theories from observations.