Liuliu writeup of Cohen 2000 KDD

This is a review of Cohen_2000_hardening_soft_information_sources by user:Liuliu.

This paper is about hardening soft databases which contain inconsistencies and duplications to hard databases. The proposed a method to determine co-references between entities and a model to find the most likely hard database, interpretation function given a soft database S. Due to the NP-hard problem, they learn the model following a greedy method choose the best arc at one time.