XML documents represent an integral part of the contemporary Web. Unfortunately, a relatively high number of them is affected by well-formedness errors, structural invalidity or data inconsistencies.
The purpose of this paper is to continue with our previous work on a correction model for invalid XML documents with respect to schemata in DTD and XML Schema languages. Contrary to other existing approaches, our model ensures that we are always able to find all minimal repairs.
The contribution of this paper is the description and experimental evaluation of our new incremental algorithm, which is able to efficiently follow only perspective correction ways even to the depth of the recursion.